Brilliant meme. Irony to the max. Is irony what makes a meme go viral? Because truth.

You can't force irony. Some people seem to have a natural gift for it. I don't. But I can recognize and appreciate it.

LESSON: If you have the gift of IRONY, learn to meme! Or team up with a memefag.

How does a meme begin in the mind of a memefag?

You might start with text in mind (in this case, a specific Q drop) and then go looking for the right image.

I often work in reverse: find an image, and then seek the right text.

I probably collected the poster (1) during an image collecting expedition (we'll talk about those later). The colors evoked the American flag, and the patriot standing in the middle obviously evoked the word "stand". Rocky Balboa in the movie, a boxer. Fight fight fight! I probably searched a text file of Q's drops for "stand" or maybe "fight", and found that perfect crumb.

(2) Erased the original text and substituted the crumb.

(How to erase? Pick the background color, in this case white, set a brush to wide, and paint over it. Then add a text layer.)

Others are free to post their ideas and observations and examples!

To Watermark? Or Not to Watermark?

Watermarks are not in the spirit of anon. I'm not here to argue copyright law, but if an image is significantly modified - say by overlaying some text - then a reasonable argument can be made that it's a different image not covered by the original copyright. Some organizations aggressively protect their copyrights and I'm not interested in pissing them off so I will stay away from their images (example: Disney cartoons, Dr. Seuss).

If somebody contributes a great meme, I have been known to remove the watermark because it's not cool to give free advertising to someone we may or may not agree with. Or you find a base image for a meme that is marked with a signature or website name.

So how do you wipe off the watermark?

Here's how I do it using the open-source free graphic editor gimp.

• Select the clone tool.

• Select a brush with soft edges.

• Enlarge the brush to a size that will cover the area you want to fix

• Select a nearby area with the graphic content you want to use, by CTRL-click

• Place your large soft clone brush over the logo or watermark, and drag left or right. The clone tool copies from the selected area, onto the location where you click & drag (i.e. the text you want to replace).

Some images are very amenable to this technique and some others are not.

1. Original image. I'm going to remove the text across the top line.

2. Identifies the Clone Tool, Soft Brush, Brush size set to 58 px. 58 px establishes the diameter of the dotted circle up top that I will copy from. Select the area to clone from by CTRL-click. (I clicked on the dotted circle.)

3. Place the cursor over the left side of the text. Click and drag right, copying from the clone source to the clone destination.

Look pretty good already.

1. Touch-up needed? Switch to the Smudge tool and enlarge the brush if needed. Drag the Smudge tool around a bit to soften the cloned area. Now you can insert new text. I picked a color from the light green text below. Done.

Now that was easy, wasn't it?

LESSON: Use the Clone Tool to remove unwanted watermarks or text from an image.

>>753

So we have concluded it really is necessary that memes for Twitter be the right SHAPE. Browbeating memefags to comply with a shape requirement hasn't worked. So a workaround is needed.

Below, we post code for a Linux command line. It would be straightforward to do the same with Windows but anon does not touch Windows (anymore). This code is simplistic - codefags could probably write something more elegant - but it works.

~~Note: 8kun does not handle files named *.jpeg correctly. Hover over back-linked .jpeg files does not work. Therefore, it has been a longstanding practice here to rename all .jpeg files to .jpg. There is no difference in a file named .jpg or .jpeg; the only difference is the file name.~~ [Removed - the .jpeg hover bug has been fixed.]

.jpg and .png file formats are fundamentally different. Renaming a .png file to .jpg or vice versa will cause problems because the filename will not correctly describe the content. Don't do this.

In these days of fast CPUs and large storage devices, why do we care about reducing file size?

Because smaller files put less of a burden on limited devices and limited bandwidth. Smaller files upload and download faster, and require less storage space. To an archivist or anyone who wants to maximize the spread of our content, reducing file sizes is common sense.

A tool to reduce image file sizes:Trimage image compressor https://trimage.org

Trimage losless image compressor - A cross-platform tool for losslessly optimizing PNG and JPG files for web. Trimage is a cross-platform GUI and command-line interface to optimize image files for websites, using optipng, pngcrush, advpng and jpegoptim, depending on the filetype (currently, PNG and JPG files are supported).

On .png files, Trimage thinks long and hard, and tries many different options until the best one is found. CPU-intensive. Reduction can vary from 0% to 50% or so, typically 5-10%. How much depends on how the graphic was created. If the file contained steganography, I presume Trimage removes it.

For .jpg files, all Trimage does it remove unnecessary color profiles, empty palette space, etc. That can reduce file size from 0% to 25% or so. Trimage works very fast on .jpg files since it has less work to do.

[edit 6-14-2020 for Twitter Shape change]

CAUTION

All these utilities work on the files in the current directory. Be careful - do not execute in your home directory. There is no "Are you sure? Y|N" prompt.

Rename all .jpeg .JPEG to .jpg

#!/bin/bash# jpeg2jpg.sh# Rename all .jpeg .JPEG to .jpgfor file in .jpeg; do mv "$file" "${file/.jpeg/.jpg}"; donefor file in .JPEG; do mv "$file" "${file/.JPEG/.jpg}"; done

# Convert all .png to .jpg

#!/bin/bash# png2jpg.sh# Convert all .png to .jpg# Some .png files should not be converted.# SVG vector graphics, .png with transparent background, .png clipart (drawings) should generally not be converted.# Images with highly detailed small text are usually best left as .png# Depends on the package ImageMagick https://imagemagick.org# This well-established graphics package should be available in the package manager for your Linux distribution# Quality of 100 (zero compression) produces extremely large file sizes.# A quality setting of 95 was selected after much experimentation. Lower quality = smaller file size. # Quality of 95 is practically imperceptible, and reduces the file size by applying a little bit of compression.# Convert all .png and .PNG to .jpg with quality of 95mogrify -format jpg -quality 95 .pngmogrify -format jpg -quality 95 .PNG# Delete those .png and .PNG filesrm .pngrm .PNG# For all files named .jpeg and .JPEG, rename them .jpgfor file in .jpeg; do mv "$file" "${file/.jpeg/.jpg}"; donefor file in .JPEG; do mv "$file" "${file/.JPEG/.jpg}"; done

Convert images to Twitter dimensions 1200 x 675 with BLACK background

#!/bin/bash# shrinkBlack.sh# Convert images to Twitter dimensions with BLACK background as needed# Execute this inside the subdirectory where the images are# Preserves original images; creates a 2nd set that are resized and renamed.# 6-14-2020 Change Twitter Dimension from 1200x627 to 1200x675 (16:9 aspect ratio)# First rename all .jpeg, .JPEG, .JPG to .jpgfor file in .jpeg; do mv "$file" "${file/.jpeg/.jpg}"; donefor file in .JPEG; do mv "$file" "${file/.JPEG/.jpg}"; donefor file in .JPG; do mv "$file" "${file/.JPG/.jpg}"; done# and rename .PNG to .pngfor file in .PNG; do mv "$file" "${file/.PNG/.png}"; done# Make backup of each file before resizingfor file in .jpg; do cp "$file" "${file/.jpg/.jpgbak}"; donefor file in .png; do cp "$file" "${file/.png/.pngbak}"; done# Resize images that are larger than 1200 pixels wide or 675 high# mogrify command provided by Imagemagick package https://imagemagick.org# This well-established graphics package should be available in the package manager for your Linux distributionmogrify -resize 1200x675 -extent 1200x675 -background black -gravity center .jpgmogrify -resize 1200x675 -extent 1200x675 -background black -gravity center .png# Append random suffix 0-9 to all .jpg and .png in current directoryfor i in .jpg; do mv "$i" "${i%.}"$((RANDOM %10)).jpg; donefor i in .png; do mv "$i" "${i%.}"$((RANDOM %10)).png; done# Restore input files to their original filenamesfor file in .jpgbak; do mv "$file" "${file/.jpgbak/.jpg}"; donefor file in .pngbak; do mv "$file" "${file/.pngbak/.png}"; done

Convert images to Twitter dimensions 1200 x 675 with WHITE background

(same as above, except substitute WHITE in place of BLACK)

# Append a random suffix 0-9 to all filenames in a directory

#!/bin/bash# appendRandomSuffix.sh# Append a random suffix 0-9 to all filenames in a directory# Leaves filename extension unchangedfor i in .jpg; do mv "$i" "${i%.}"$((RANDOM %10)).jpg; donefor i in .png; do mv "$i" "${i%.}"$((RANDOM %10)).png; done

>>753

>>1149

>>340

Memefags read this.

For 2020, the recommended image size for twitter is now a 16:9 aspect ratio.

https://louisem.com/217438/twitter-image-size

For tweeted images:

Image width/height: Minimum 600 X 335 pixels, although larger images (for example 1200 X 675) will be better optimized for when users click to expand images.

Aspect ratio: 16:9.

Image file size: Max 15mb on twitter.com and 3mb on ads.twitter.com.

File types: PNG, JPEG, or GIF are recommended. Twitter does not accept BMP or TIFF files.

As of October 2109, images taller than this 16:9 proportion will be cropped in the feed on both mobile and on desktop – except for GIFs and videos, which can appear up to square.

It is recommended to focus on mobile, since 85% of images displayed on Twitter are displayed on a mobile.

• • • • • • • • • -

Therefore our recommended size/shape for twitter is now 1200 x 675 (instead of 1200 x 600). That shape will be a lot easier for memefags to work with.

I hope others will chime in. Here's how I go about locating images for memes.

• Primary tool: search engine. Try various search strings. ~~Personally I always use Duck Duck Go and never Google Images, unless DDG completely strikes out.~~ The Russian search engine

https://yandex.com

has far more and better images than Duck Duck Go. I rarely use Google Images, only as a last resort. Search for your string, then click Images to see a page of images only. Set the size (usually Large) to filter out low-resolution images. I will sometimes try a dozen search strings to zero in on what I'm looking for. Sometimes I don't know what I'm looking for, but perusing a trail of image breadcrumbs generates ideas and leads to interesting content. Google's database is more comprehensive, but it features left-leaning content and left-leaning categories, diminishing its usefulness (to me). Google has a nice feature to search for images resembling another image. This can sometimes get the creative juices flowing, if you're feeling stuck.

• Avoid images that are watermarked – because it's a pain to fix them. May have to add filters to search string to exclude commercial image repositories: -dreamstime -stock -alamy etc.

• Use images from current news events

• Make collections of images for future use and file them under Batter. Scan your own image collection from time to time; it will suggest meme ideas.

Other ideas

– deepdreamgenerator.com has interesting images

–pxhere.com royalty-free amateur photos; some are really good

– powerpoint templates

– fantasy art, angels, fantasy heroes, fantasy landscape, science fiction art, space art

– Pinterest (no you don't need an account to peruse their images; you just have to reject the login prompt by clicking the invisible X in the upper right corner of the pop-up)

– Superheroes, comics, and anime

– Deviantart.com, has a lot of options to filter for certain kinds of content

– Wallpaper sites (there are many)

– Posters, silk-screen art, travel posters, vintage posters

– Vector drawings

– Web design and commercial art sites for color and palette ideas

– Landscape photography

– Candid photos (filter out -wedding -India)

– Youtube videos: use screencapture to pick a still image out of a video)

– Old TV shows

Vary the search by filtering for transparent images, clip art, gif animations, etc.

The same search will locate different content at a future time.

If too many images come up, keep adding search terms to narrow it. E.g. smiling woman -smiling woman face -> smiling black woman face -> smiling black woman gesture etc.

Try your search string in quotes, or without quotes, or quotes around a phrase within a longer string, to vary the results.

Be persistent. Plan to spend time finding images.

While looking for one kind of image, you will likely encounter other images that you don't need right now, that could be useful later. Throw them into that Batter folder.