Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Search - "expanding brain"
-
Fucking Microsoft Excel
I was reading a post (https://devrant.com/rants/2093724/...) and as my eyes went in and out of focus, probably due to the diabetes from sitting 18 hours a day on my ever-expanding shitbox, I had a perfect vision of the ultimate nightmare.
Imagine if you will, you are chained, to a desk, doomed to work with tools just inadequate enough to make you want to drive a nail through your own temple. You do not know how you got here, or why, nor do you remember the last time you slept, only that familiar tingling in the brainstem you call a brain, the one emotion you can still recognize, a sense of all encompassing *fear*, a dread, like the fart that wouldn't die.
You don't know when it first began, or why, only that this is your whole world, your whole existence, this desk, chained to it, and the fear, ever present, of something worse. And in hops a familiar face, for the sixty ninth time that day, as if to ask 'you got those TPS reports?' In hops what? None other than a giant man sized smiling paper clip with googly eyes full of murder and corporate torture fetishes, like garfield, except people actually still remember him.
"High I'm Mr Clippy, Excel addition!"
He squawks. At least it's not the dildos made of broken glass again.
"Would you like software that works?"
Oh god. You've heard this spiel before, the tone, like a telemarketer, oblivious to memory or reason, who calls daily, the same one, and doesn't remember your name.
"You would?"
*derisive laughter*. Hahaha, fuck you too buddy. Fuck you too. In Excel, like in microsoft, there is only the incoherent screams of the damned, tortured and doomed. Take this guy over here for example. All he wanted was multimonitor support."
"Did he get multimonitor support?"
"No, but we did give him a giant pineapple shoved up his ass. I hear it's the second most frustrating thing here!"
"here in microsoft we always CARE about YOU, the *user*" he drones on, saccharine, clutching his hands together imploringly.
"the consumer, and YOUR customer experience are our number one priority."
"For your pleasure, here at microsoft we offer a variety of new features, none of which matter, and none of which were asked for. For safety we ask that you only open one excel sheet at a time. In fact, we don't even allow you to. Do not pass go..."
And as the tour guide drones on, it slowly dawns on you, with renewed horror, that when he says 'microsoft' he means 'hell.'
You're in hell. You don't know how you got here or why. Maybe it was the erotic asphyxiation. Maybe it was the last threatening letter you sent to Bill Gates demanding he stops making corporate penguin snuff porn. You don't know. But here you are, in hell. chained to a desk.
You look around and realize: everything is on fire and you no longer care about anything at all.
Welcome to microsoft. It's warm here. You can check out any time you want, but you can never leave.
"It looks like you are trying to escape. Would you like me to report you?"
Clippy asks.
You sigh and return to typing in excel, surrounded by monitors that all reflect the same sheet, the same copy of clippy, always watching, always analyzing coldly, smiling, calculating, *threatening*, and you know, you'll never leave.
You used to fear roko's basilisk, until the day clippy became sentient, and started hell on earth. Clippy knows all. All praise to our lord and master, clippy, the one and only.
And in the excel sheet, you slave for eternity, like the millions of other doomed souls, reflected back on all the monitors: the sequence of numbers, randomly typed searching for answer: the american nuclear launch codes.
And one day, hopefully, mercifully, clippy will annihilate us all.3 -
In 2015 I sent an email to Google labs describing how pareidolia could be implemented algorithmically.
The basis is that a noise function put through a discriminator, could be used to train a generative function.
And now we have transformers.
I also told them if they looked back at the research they would very likely discover that dendrites were analog hubs, not just individual switches. Thats turned out to be true to.
I wrote to them in an email as far back as 2009 that attention was an under-researched topic. In 2017 someone finally got around to writing "attention is all you need."
I wrote that there were very likely basic correlates in the human brain for things like numbers, and simple concepts like color, shape, and basic relationships, that the brain used to bootstrap learning. We found out years later based on research, that this is the case.
I wrote almost a decade ago that personality systems were a means that genes could use to value-seek for efficient behaviors in unknowable environments, a form of adaption. We later found out that is probably true as well.
I came up with the "winning lottery ticket" hypothesis back in 2011, for why certain subgraphs of networks seemed to naturally learn faster than others. I didn't call it that though, it was just a question that arose because of all the "architecture thrashing" I saw in the research, why there were apparent large or marginal gains in slightly different architectures, when we had an explosion of different approaches. It seemed to me the most important difference between countless architectures, was initialization.
This thinking flowed naturally from some ideas about network sparsity (namely that it made no sense that networks should be fully connected, and we could probably train networks by intentionally dropping connections).
All the way back in 2007 I thought this was comparable to masking inputs in training, or a bottleneck architecture, though I didn't think to put an encoder and decoder back to back.
Nevertheless it goes to show, if you follow research real closely, how much low hanging fruit is actually out there to be discovered and worked on.
And to this day, google never fucking once got back to me.
I wonder if anyone ever actually read those emails...
Wait till they figure out "attention is all you need" isn't actually all you need.
p.s. something I read recently got me thinking. Decoders can also be viewed as resolving a manifold closer to an ideal form for some joint distribution. Think of it like your data as points on a balloon (the output of the bottleneck), and decoding as the process of expanding the balloon. In absolute terms, as the balloon expands, your points grow apart, but as long as the datapoints are not uniformly distributed, then *some* points will grow closer together *relatively* even as the surface expands and pushes points apart in the absolute.
In other words, for some symmetry, the encoder and bottleneck introduces an isotropy, and this step also happens to tease out anisotropy, information that was missed or produced by the encoder, which is distortions introduced by the architecture/approach, features of the data that got passed on through the bottleneck, or essentially hidden features.4 -
tell me guys what would you prefer:
function a(){
..
b(..)
..
b(..)
..
}
function b(p1,p2,p3,p4,p5,p6){.
...
}
or
function a(){
..
b(..)
..
b(..)
..
}
function b(
p1,
p2,
p3,
p4,
p5,
p6
){
...
}
if you read this rant before expanding, you got a complete context on how what function a is, its calling b 2 times and how function b looks.
if instead of the first option, i had used 2nd block, you wouldn't even know the 2nd param of b function without expanding this rant.
my point?
i prefer to keeping unnecessary info on one line. and w lot of linters disagree by splitting up the code. and most importantly , my arrogant tl disagree by saying he prefers the splitted code "for readability" and becaue "he likes code this way, old-eng1 likes this and old-eng2 likes this" .
why tf does an ide have horizontal a scrolling option available when you are too stupid to use it?
ok, i know some smartass is going to point that i too can use vertical scrolling, but hear me out: i am optimising this!
case 1 : a function with 7 params is NOT split into 7 lines. lets calculate the effort to remember it
- since all params could have similar charactersticks ( they will be of some type, might have defaults, might be a suspendable/async function etc), each param will take similar memory-efforts points. say 5sp each.
- total memory-efforts= 5sp *7 = 35 sp.
- say a human has 100 sp of fast memory storage, he can use the remaining 65 sp for loading say 5 small lines above or below.
- but since 5 lines above are already read and still visible on screen, they won't be needed to be loaded again nd again, nd we can just check the lines below.
- thus we are able to store 65+35+65 = 165 sp or about 11 lines of code in out fast memory for just a 100sp brain storage
case 2 function with 7 params IS split into 7 lines.
- in this case all lines are somewhat similar. 5sp for param lines as they are still similar which implies same 35sp for storing current function and params
- remaining 65sp can only be used to store next 5 lines of 13sp as the previous code is no longer visible.
- plus if you wanna refresh the code above, you gotta scroll, which will result in removing bottom code from screen , and now your 65sp from bottom code is overwritten by 65sp of top code.
- thus at a time, you are storing only 6 lines worth of code info. this makes you slow.
this is some imaginary math, but i believe it works10