2080

Yeah, I didn’t know it either.

I’m sort of tickled by the idea that an extra 144 beers would require a call back to prevent. Ohhhhh, we can’t let that happen! Such lightweights.

2081

2082

2083

image
image.jpg996×666 74 KB

2084

My first reading of that was very wrong. It’s kinda worded weird.

2085

The ‘classic headline writer goof’ is what makes it interesting (to me, at least).

https://www.amazon.com/Eats-Shoots-Leaves-Tolerance-Punctuation/dp/1592402038?tag=qt3-20

2086

Got it. N word stadium would be a very bad idea. :)

2087

image
image.jpg612×612 107 KB

2088

2089

2090

I just watched “Brainstorm” on TCM, and saw this in the credits.

MVIMG_20180721_194619
MVIMG_20180721_194619.jpg1511×2015 1.42 MB

2091
2092

I want to watch this.

2093

The “Well, I’m the government now” line by The Kramer, had me howling. I can already imagine George being exasperated by a job-stealing bird.

2094

The Olive Garden one is inspired, for those who haven’t seen it.

I assume he’s doing a bunch of these now.

2095

“There’s more Italy than necessary” has become a catch phrase at our house.

2096

I love the Seinfeld one, and “more Italy than necessary” is a great line too.

One really pedantic nitpick is that I don’t like that this guy describes it as “I forced a bot to watch over 1000 hours of X”. For a machine learning algorithm (I presume a neural network here) to actually watch 1000 hours of anything it would be converting the frame images to pixel information and then processing it. If it truly watched the episodes or ads it could try to generate new episodes / ads visually. It’s actually reading 1000 hours of episode scripts. The word ‘forced’ has no meaning here either.

Moreover, there is less than 90 hours of Seinfeld, so I have no idea where 1000 comes from.

I also wonder with this ML generated literature how much forcing or cheating is going on. Some things just seem too perfect, and I think it would be very tempting for a person producing these to fiddle with the words here and there so the punchlines give maximum effect. Not claiming this particular guy is doing it but just a thought.

2097

I assume he means he fed it the scripts for X shows.

2098

Hate to ruin the joke, but those are 100% human-written humor. Actual ML generation of relatively simple things like lists of name, sure, but not full scripts.

2099

So the guy is a comedian and after thinking about it it’s pretty obvious it’s just a joke on other people doing this machine learning generated stuff. None of “I forced a bot to watch over 1,000 hours of Seinfeld” makes any sense.

In any event, it was funny! Props to him.

EDIT: See above, now I feel slow. To make matters worse, my post appears after Thraeg’s so there is no proof that I came to this conclusion on my own! Curses!