WorstPlans.com updates every Monday!

Your weekly source for terrible plans and ideas!

Tag: machine learning

Stop getting hit by self-driving cars with this one fashion trick that involves putting weird labels on all your clothing! Don’t be the last one to catch on to this new fashion trend.


In a hypothetical future where self-driving cars are increasingly common, they’ll have to do a really good job of automatically distinguishing between things that require sudden braking (e.g. a person in the roadway) and things that are OK to hit (e.g. a tumbling empty cardboard box).

The issue:

This is a hard problem. When a car gets data from its various cameras (and other sensors), it needs to figure out what exactly it is that it is seeing (Figure 1).


Fig. 1: This is probably a pedestrian in the roadway, but could it also be a billboard advertisement hundreds of feet away?

Although the specific “distant-billboard-or-close-pedestrian” question in Figure 1 can be answered just by using two cameras to estimate distance, there are situations where the problem must be resolved in a more complex fashion (Figure 2).


Fig. 2: Top: the image is interpreted correctly, and the car does NOT hit the pedestrian. Bottom: the car incorrectly believes that it sees a sunflower, and collides with it at full speed. Lest you think this is totally implausible, check out some specially-crafted adversarial examples (that can turn a panda into a banana) and a method of tricking lane-following algorithms into swerving the car into oncoming traffic.


We propose to place special “this is a human” symbols on articles of clothing that a human might wear (Figure 3).

When a car sees one of these unusual QR-code-like symbols, it will instantly say “ah, sunflowers do not wear specially-marked shoes, time to hit the brakes!”

To avoid this becoming a fashion disaster, these markings would not be apparently at normal human-visible wavelengths of light, but would only be detectable by special camera equipment.

Perhaps the markings could have fluorescent ink in them, and all cars could drive around with UV lights in the front.


Fig. 3: Left: this is what the shoe looks like to a human—the markings are invisible to the naked eye. Middle: the camera can see wavelengths of light beyond human ability, and can detect these special markings (shown here as yellow checkerboards). Right: the camera sees the checkerboard, and the object-classification algorithm realizes that this shoe is likely to be attached to a human.

One common objection to many self-driving-car-related issues is “couldn’t some criminal put these markers all over the city, to trick self-driving cars?”

The answer is yes, but it would be as equally illegal as it currently is to put mannequins on a winding road (which would also confuse human drivers).


This might be redundant with an infrared camera—in most locations, a human already is obviously distinguished from the background environment just by their warm-blooded glow in the infrared spectrum.

PROS: This will definitely make me a ton of money when it is licensed by major car manufacturers. Also, would someone please apply for and pay for a patent on my behalf? Thanks!

CONS: If one of these specially-marked shoes falls onto the roadway (perhaps by falling out of someone’s messenger bag while they’re biking), do we really want every car to come to a screeching halt at the sight of a single unattached shoe?



The secret of SMART JUSTIFIED columns of text. This strange formatting tip will make ONE HUNDRED TIMES more employers look at your resume! Stop formatting your resume so amateurishly, and await your reward of gold and rubies from your future employer.


Columns of text in a book or newspaper are generally formatted in the fully justified style (Figure 1), where the text always lines up exactly on both the left and right edges.


Fig. 1: The “justify text” button (circled in red) can be found in nearly every text editor.

The issue:

Justified text works well if columns are wide and there are a lot of words to fill out each line.

But it becomes aesthetically dubious if the columns are narrow or there aren’t enough words, which result in either:

  • Extremely wide spaces between words if there are too few words (example: “this______column”)


  • Excessive spacing between letters if there is only one word (example: “c__o__l__u__m__n”)

In the worst-case scenario, a column of text may look like:

  • This____is_____a
  • n__a__r__r__o__w
  • c__o__l__u__m__n.

See figure 2 for a comparison of fully-justified text and ragged-edge (flush left) text.


Fig. 2: Part A (left) shows a few problems with fully-justified text: “the age of” has excessive spacing and the between-letter spacing in “w i s d o m” is aesthetically questionable. Unfortunately, the ragged edge of the text in part B (formatted as “flush left / ragged right”) is not a huge improvement either.

Previously, a publisher would at least know how wide a column of text would be, so they could manually adjust the text to fit in an aesthetically-appealing fashion.

But with modern web pages and e-books, font sizes and column widths can be changed by the user—so there’s no way for a publisher to plan around it.


This problem can be fixed by using semantically-aware SMART justification to make each line of text an optimal length.

This is accomplished as follows:

If a line of text is too short, it can be lengthened by the following steps:

  • Add meaningless filler words (e.g. “um,” “like,” “basically,” “you know”)
  • Add superfluous adjectives (like “very” or “extremely”)
  • Replace words with longer synonyms (e.g. “rain -> precipitation”—this can also be used in reverse to shorten a line)
  • Replace pronounces with their antecedent (e.g. “her scepter” -> “Queen Elizabeth’s scepter”)

Figure 3 shows the performance of each method of text justification. The “meaning-aware SMART justification” is the only method that avoids ragged edges while also keeping a fixed amount of whitespace between words.


Fig. 3: Left: a traditional example of fully-justified text. Middle: flush-left text, with an unappealing ragged right edge. Right: the vastly improved “smart” justification method, which has been recently made possible by advances in computational technology and machine learning.

Application of this method to famous books:

  • Original: “But man is not made for defeat,” he said. “A man can be destroyed but not defeated.” (The Old Man and the Sea, Hemingway)
  • Modified with superfluous filler words and synonyms:  “But man is, generally, not made for defeat,” he stated. “Basically, a man can be destroyed but, as you know, not forced to surrender.” 


  • Original: “War is peace. Freedom is slavery. Ignorance is strength.” (1984, Orwell)
  • Modified:  “War is peace. Additionally, the state of freedom is slavery. Finally, in conclusion, ignorance is strength, it must be admitted.”


  • Original: “In general, people only ask for advice that they may not follow it; or, if they should follow it, that they may have somebody to blame for having given it”.” (The Three Musketeers, Dumas)
  • Modified: “In general, people only make a request for suggestions, that those same people may not abide by it. Or, if they should in fact follow it, that those people may have somebody to blame or hold responsible for having given it”.” 


PROS: This is the ONLY text-formatting method that both 1) preserves inter-word spacing AND 2) aligns text in neat columns.

CONS: None!