Yet another commercial seminar upset I won't give up a day of my time to fly to the UK to speak at the event for no payment or reimbursement for my travel.

But of course I should think about the exposure and networking opportunities! 😕

Base10 is interesting.
Did you know that any number > 9 has this weird thing where if you sub it's digits from the number over and over, it'll eventually become 9?

For example: 35
35 - 3 - 5 = 27
27 - 2 - 7 = 18
27 - 1 - 8 = 9

Try it with any number you like.

Aint a rant though
But I woke up today and the first thing I saw was my first Android app reaching 5 digits (10017) downloads figure with 4.73 stars. :D
Couldnt be happier.

I met this girl on the way back from lunch, I stopped, requested her digits, when I asked "what's your address" and she replied "123.2.2.8". I fell in love instantly. I knew she was the one.

My Neural Network can recognise handwritten digits!!!

It's my second try at NN so it's faaaar from perfect (or maybe even good), but hey, it's something and only with High schools, I'm pretty satisfied with the results. If you've not seen my previous post, I'm just trying to learn NNs in C and am doing just really basic things.
Still I'm proud of my progress!

Now I'm looking forward learning some library (OpenNN + OpenCV seems cool) and trying more advanced stuff, wish me luck 😆

After one job interview I ran across this one guy in an elevator.

"Are you an IT? I am looking for one".

I nodded, "Yes, I am". He was about to get off the elevator and told me to save his phone number.

By the time I was about to type the last 3 digits of his number on my mobile, I got a call from my brother and the elevator door closed. I immediately rejected the call.

Unfortunately, the dialer went empty and I lost his number.

I was trying to recall the number but I can only remember the last 3 digits.

I went back to the same floor, but there was around 30 offices and I couldn't find him. I gave up.

Fucking iPhone.

Biggest scaling challenge I've faced?

Around 2006~2007 the business was in double-digit growth thanks to the eCommerce boom and we were struggling to keep up with the demand.

Upper IT management being more hardware focused and always threw more hardware at the problem. At its worst, we had over 25 web servers (back then, those physical tall-rectangle boxes..no rack system yet) and corresponding SQL server for each (replicated from our main sql server)

Then business boomed again and projected the need for 40 servers (20 web servers, 20 sql servers) over the next 5 years. Hardware+software costs (they were going to have to tear down a wall in order to expand the server room) were going to be in the $$ millions.

Even though we were making money, the folks spending it didn't seem to care, but I knew this trajectory was not sustainable, so I started utilizing (this was 2007) WCF services and Microsoft's caching framework Velocity. Started out small, product lookup data (description, price, the simple stuff) and within a month, I was able to demonstrate the web site could scale with less than half of our current hardware infrastructure.

After many political battles (I've ranted about a few of those), the $$ won and even with the current load, we were able to scale back to 5 web servers and 2 sql servers. When the business increased in the double-digits again, and again...we were still the same hardware for almost 5 years. We only had to add another service server when the international side of the business started taking off.

Challenge wasn't the scaling issue, the challenge was dealing with individuals who resisted change.

We call “Binary digITS” “bits”.

So then what would we call ternary digits?

What do you get distracted from?

How the fuck does these testers find these bugs
there are like
- go to this
- click on this
- type the first 57 digits of pi
- sacrifice your first born to the devil
-> there the table is misaligned by 13 pixels, fix that

I'd rather memorise all the digits of PI than look at your code

In a programming contest, I forgot how to round numbers in Java, an I needed a 3 number rounding, so I multiplied the number by 1000, then sum 0.5 and convert it to integer so the decimal part would be gone, finally, just print the number except the 3 last digits as a string, put a period and print the other 3 digits.
I must say I'm not proud of that.

some miserable life...

I messaged a professor at MIT and surprisingly got a response back.

He told me that "generating primes deterministically is a solved problem" and he would be very surprised if what I wrote beat wheel factorization, but that he would be interested if it did.

It didnt when he messaged me.

It does now.

Tested on primes up to 26 digits.
Current time tends to be 1-100th to 2-100th of a second.

Seems to be steady.

First n=1million digits *always* returns false for composites, while for primes the rate is 56% true vs false, and now that I've made it faster, I'm fairly certain I can get it to 100% accuracy.

In fact what I'm thinking I'll do is generate a random semiprime using the suspected prime, map it over to some other factor tree using the variation on modular expotentiation several of us on devrant stumbled on, and then see if it still factors. If it does then we know the number in question is prime. And because we know the factor in question, the semiprime mapping function doesnt require any additional searching or iterations.

The false negative rate, I think goes to zero the larger the prime from what I can see. But it wont be an issue if I'm right about the accuracy being correctable.

I'd like to thank the professor for the challenge. He also shared a bunch of useful links.

That ones a rare bird.

I swear to god, I'm going to track down the dipshit who just made my day hilariously painful.

So here I am, finishing up this project that's been going on for what feels like an eternity, when I get an email "why doesn't order X show up in this other system?".

I mean, it's a common thing they can take 15 minutes to push across, so the usual quick glance and what do you know, it's just sitting there as if it's waiting to be pushed through, than an hour later... it's still there, so I start digging, maybe a data issue, nope looks all good, customer details, payment details, products...

just another order, jump on the logs and all looks fi......... wait.... why does this postcode have 3 digits and not 4 , Australia has 4 digit postal codes fyi, looks at order again, 3 digits, look at log, 3....hold on why's it only 3 digits, checks code, handled as string... ok..... where the fuck would it drop a digit.... frontend requires 4 digits, validation requires 4 digits... how the fuck did you get 3 digits in... I can't see anything anywhere that logically makes sense for this🤔

Drops address into google and it's a postcode starting with 0.

Jumps on DB and the fucker is an int in the postcode table. For all you playing at home 0123 <> 123

I don't know if I should feel bad, or impressed, it's been 7 years since this table was created, and 7 years before someone managed to live in one of these parts of the country with a leading 0.

QA didn't spot this years ago,
No one tested this exact scenario,
The damn thing isn't even documented as a required delivery area, but here we are!

Kudos good sir, you broke it! 🤜 🤛
You sir may get your order now!

POSTMORTEM

"4096 bit ~ 96 hours is what he said.

IDK why, but when he took the challenge, he posted that it'd take 36 hours"

As @cbsa wrote, and nitwhiz wrote "but the statement was that op's i3 did it in 11 hours. So there must be a result already, which can be verified?"

I added time because I was in the middle of a port involving ArbFloat so I could get arbitrary precision. I had a crude desmos graph doing projections on what I'd already factored in order to get an idea of how long it'd take to do larger
bit lengths

@p100sch speculated on the walked back time, and overstating the rig capabilities. Instead I spent a lot of time trying to get it 'just-so'.
Worse, because I had to resort to "Decimal" in python (and am currently experimenting with the same in Julia), both of which are immutable types, the GC was taking > 25% of the cpu time.

Performancewise, the numbers I cited in the actual thread, as of this time:
largest product factored was 32bit, 1855526741 * 2163967087, took 1116.111s in python.
Julia build used a slightly different method, & managed to factor a 27 bit number, 103147223 * 88789957 in 20.9s,
but this wasn't typical.

What surprised me was the variability. One bit length could take 100s or a couple thousand seconds even, and a product that was 1-2 bits longer could return a result in under a minute, sometimes in seconds.

This started cropping up, ironically, right after I posted the thread, whats a man to do?

So I started trying a bunch of things, some of which worked. Shameless as I am, I accepted the challenge. Things weren't perfect but it was going well enough. At that point I hadn't slept in 30~ hours so when I thought I had it I let it run and went to bed. 5 AM comes, I check the program. Still calculating, and way overshot. Fuuuuuuccc...
So here we are now and it's say to safe the worlds not gonna burn if I explain it seeing as it doesn't work, or at least only some of the time.

Others people, much smarter than me, mentioned it may be a means of finding more secure pairs, and maybe so, I'm not familiar enough to know.

For everyone that followed, commented, those who contributed, even the doubters who kept a sanity check on this without whom this would have been an even bigger embarassement, and the people with their pins and tactical dots, thanks.

So here it is.
A few assumptions first.
Assuming p = the product,
a = some prime,
b = another prime,
and r = a/b (where a is smaller than b)
w = 1/sqrt(p)
(also experimented with w = 1/sqrt(p)*2 but I kept overshooting my a very small margin)

x = a/p
y = b/p

1. for every two numbers, there is a ratio (r) that you can search for among the decimals, starting at 1.0, counting down. You can use this to find the original factors e.x. p*r=n, p/n=m (assuming the product has only two factors), instead of having to do a sieve.

2. You don't need the first number you find to be the precise value of a factor (we're doing floating point math), a large subset of decimal values for the value of a or b will naturally 'fall' into the value of a (or b) + some fractional number, which is lost. Some of you will object, "But if thats wrong, your result will be wrong!" but hear me out.

3. You round for the first factor 'found', and from there, you take the result and do p/a to get b. If 'a' is actually a factor of p, then mod(b, 1) == 0, and then naturally, a*b SHOULD equal p.
If not, you throw out both numbers, rinse and repeat.

Now I knew this this could be faster. Realized the finer the representation, the less important the fractional digits further right in the number were, it was just a matter of how much precision I could AFFORD to lose and still get an accurate result for r*p=a.

Fast forward, lot of experimentation, was hitting a lot of worst case time complexities, where the most significant digits had a bunch of zeroes in front of them so starting at 1.0 was a no go in many situations. Started looking and realized
I didn't NEED the ratio of a/b, I just needed the ratio of a to p.

Intuitively it made sense, but starting at 1.0 was blowing up the calculation time, and this made it so much worse.

I realized if I could start at r=1/sqrt(p) instead, and that because of certain properties, the fractional result of this, r, would ALWAYS be 1. close to one of the factors fractional value of n/p, and 2. it looked like it was guaranteed that r=1/sqrt(p) would ALWAYS be less than at least one of the primes, putting a bound on worst case.

The final result in executable pseudo code (python lol) looks something like the above variables plus

while w >= 0.0:
if (p / round(w*p)) % 1 == 0:
x = round(w*p)
y = p / round(w*p)
if x*y == p:
print("factors found!")
print(x)
print(y)
break
w = w + i

Still working but if anyone sees obvious problems I'd LOVE to hear about it.

Lets get to know eachother, shall we..?

It's come to my attention that all we do is either rant, or post memes! Lets talk! D;

What's your favorite programming language?

How about spoken/written language?

Favorite band?

Song?

Do you sit or stand while working?

What's the last four digits of your SSN? Haha, jk..

What are you passonate about?

What projects (programming or otherwise) have you started/picked up/finished lately?

What else do you want to share?

Answer as many as you'd like~

What a fucking day!
So some many years ago, we placed a small and useful piece of code into a MySQL Query, which prefixes the order ids with a string and returns the value.

Basically

SELECT
CONCAT(“ORDER_PREFIX”, LPAD(order.id,5,’0’) AS order_reference

Now many years ago when this was a simple solution to a system that wasn’t expected to be running today, all went well, except today.... today that fucking line exceeded 5 digits
😓😍😓

I don’t know if I should be ashamed, or happy right now, a totally unfuture proof query that just rolled that little too far, or the fact that I’ve watch 100k orders role through this system.

Now for the clean up crew to do their job.... oh 🙁 that’s me too.

Now this doesn’t affect the order ids... so that’s something but external systems rely on that guy 😭

Best carrier choice: leaving the old company I worked for...

TLDR version: I yelled at owner/director to shove it..got layed off the next day. Never regreted that day!! EVER!!

Long version: I was asked to implement storing of CC numbers (yes, the whole lot) & their matchig CVC numbers..plain text, no encription.. company didn't even fulfilled standards to store last four digits.. so I yelled at the 'big boss' that he is crazy and that I won't do it.. Next day, I got handed a letter that said they have to lay me off due to lack of work and that my position is redundant & no other workplace for me.. Never been happier in my life!!

I wanted to quit for some time, due to crazy stuff they asked me to implement & how!? & toxic personel. I was called Hitler (& am actually proud of this!!) cuz I was work oriented & didn't kiss mrs.Director's ass.. I wasn't slacking like the rest of them did, so of course, I was the bad guy..

Anyhow, fast forward 6 months, got the best job ever & am now here for 5yrs+.

TL;DR :

"when i die i want my group project members to lower me into my grave so they can let me down one last time"

STORY TIME

Last year in College, I had two simultaneous projects. Both were semester long projects. One was for a database class an another was for a software engineering class.

As you can guess, the focus of the projects was very different. Databases we made some desktop networked chat application with a user login system and what not in Java. SE we made an app store with an approval system and admin panels and ratings and reviews and all that jazz in Meteor.js.

The DB project we had 4 total people and one of them was someone we'll call Frank. Frank was also in my SE project group. Frank disappeared for several weeks. Not in class, didn't contact us, and at one point the professors didn't know much either. As soon as we noticed it would be an issue, we talked to the professors. Just keeping them in the loop will save you a lot of trouble down the road. I'm assuming there was some medical or family emergency because the professors were very understanding with him once he started coming back to class and they had a chance to talk.

Lesson 1: If you have that guy that doesn't show up or communicate, don't be a jerk to them and communicate with your professor. Also, don't stop trying to contact the rogue partner. Maybe they'll come around sometime.

It sucked to lose 25% of our team for a project, but Frank appreciated that we didn't totally ignore him and throw him under the bus to the point that the last day of class he came up to me and said, "hey, open your book bag and bring it next to mine." He then threw a LARGE bottle of booze in there as a thank you.

Lesson 2: Treat humans as humans. Things go wrong and understanding that will get you a lot farther with people than trying to make them feel terrible about something that may have been out of their control.

Our DB project went really well. We got an A, we demoed, it worked, it was cool. The biggest problem is I was the only person that had taken a networking class so I ended up doing a large portion of the work. I wish I had taken other people's skills into account when we were deciding on a project. Especially because the only requirement was that it needed to have a minimum of 5 tables and we had to use some SQL language (aka, we couldn't use no-SQL).

The SE project had Frank and a music major who wanted to minor in CS (and then 3 other regular CS students aside from me). This assignment was make an app store using any technology you want. But, you had to use agile sprints. So we had weekly meetings with the "customer" (the TA), who would change requirements on us to keep us on our toes and tell us what they wanted done as a priority for the next meeting. Seriously, just like real life. It was so much fun trying to stay ahead of that.

So we met up and tried to decided what to use. One kid said Java because we all had it for school. The big issue is trying to make a Java web app is a pain in the ass. Seriously, there are so many better things to use. Other teams decided to use Django because they all wanted to learn Python. I suggested why not use something with a nice package system to minimize duplicating work that had already been done and tested by someone. Kid 1 didn't like that because he said in the real world you have to make your own software and not use packages. Little did he know that I had worked in SE for a few years already and knew damn well that every good project has code from somewhere else that has already solved a problem you're facing. We went with Java the first week. It failed miserably. Nobody could get the server set up on their computers. Using VCS with it required you to keep the repo outside of the where you wrote code and copy and paste changes in there. It was just a huge flop so everyone else voted to change.

Lesson 3: Be flexible. Be open to learning new things. Don't be afraid to try something new. It'll make you a better developer in the long run.

So we ended up using Meteor. Why? We all figured we could pick up javascript super easy.Two of us already knew it. And the real time thing would make for some cool effects when an app got a approved or a comment was made. We got to work and the one kid was still pissed. I just checked the repo and the only thing he committed was fixing the spelling of on word in the readme.

We sat down one day and worked for 4 straight hours. We finished the whole project in that time. While other teams were figuring out how to layout their homepage, we had a working user system and admin page and everything. Our TA was trying to throw us for loops by asking for crazy things and we still came through. We had tests that ran along side the application as you used it. It was friggin cool.

Lesson 4: If possible, pick the right tool for the job. Not the tool you know. Everything in CS has a purpose. If you use it for its purpose, you will save days off of a project.

Where I work, in our database, we use 3 to indicate true and 7 to indicate false and 0 is true but null is false

In another table, we use 'P' to indicate true and 'I' to indicate false and 'Y' is also false and null is false

And the most used table, we also use 'Y' to indicate yes and 'N' to indicate no, but null is also Yes.

We also store integers as varchar in a live table, but stays an integer in all the other tables. I hope I'm not there when the number of digits exceeds the varchar limit.

These are all live and used in production all created by my boss, the head of IT.

I started early in my childhood days, nobody had cellphone or internet here, my phone number was 3 digits long and my home country started to recover from 44 years of communism.

My first dev project was probably to copy game from newspaper to Atari 1300XE

Article listing was around 10 pages long and if you made mistake program didn’t run.
It took me a while I can’t remember how long but probably whole day and I was finally able to play it.

I don’t remember what was game about but later on I learned some BASIC from book and was able to color the screen and stuff like that.

I was about 6 years old.

I also remember that Atari computer had tape recorder where you put cassette to load game.
Some more complicated games were loading more then hour and you need to walk very carefully around or your walk can cause error and operation would fail.

Besides that there were national radio auditions about Atari where at the end they played code sound wave so you can record it on your cassette and then play software from radio on your Atari.

I never managed to do it cause I was living near military airport and pilots were practicing landing and starting above my home causing radio signal noise and breaking my software recoding.

I can probably say that highly accelerating plane could cause game loading problem and it’s not a joke.

So... um... I'm a real asshole... but this website where I'm applying for a job doesn't check how many digits should be entered for their GPA field.

At first, I wanted to see if I could put in the pi symbol (π). That didn't work.

So, then I added 10 digits of π into the field. When I saw that it had accepted that many numbers, I grew curious and wondered how many numbers can I fit in here.

I added 100, then 1000, then a million.

Sooooo, do you guise think I'll get the job?

me vs vmware this week

Yesterday i bought train ticket to go home and at the ATM this happened :
@wowotek : tell me the exact price nominal
@aureliagbrl : 293,000 Rupiahs (~10 USD)
Me : *entering the nominal fast and rough* 2,930,000
A : hey careful you put too much zero--
Me: *just realize but i clicked the accept button too quickly* doesn't matter my savings is only 6 digits. *re entering the correct nominal*

...

Also Broke ass me : *cries in the corner*

Life is Hard.

math be like:

"Addition (often signified by the plus symbol "+") is one of the four basic operations of arithmetic; the others are subtraction, multiplication and division. The addition of two whole numbers is the total amount of those values combined. For example, in the adjacent picture, there is a combination of three apples and two apples together, making a total of five apples. This observation is equivalent to the mathematical expression "3 + 2 = 5" i.e., "3 add 2 is equal to 5".

Besides counting items, addition can also be defined on other types of numbers, such as integers, real numbers and complex numbers. This is part of arithmetic, a branch of mathematics. In algebra, another area of mathematics, addition can be performed on abstract objects such as vectors and matrices.

Addition has several important properties. It is commutative, meaning that order does not matter, and it is associative, meaning that when one adds more than two numbers, the order in which addition is performed does not matter (see Summation). Repeated addition of 1 is the same as counting; addition of 0 does not change a number. Addition also obeys predictable rules concerning related operations such as subtraction and multiplication.

Performing addition is one of the simplest numerical tasks. Addition of very small numbers is accessible to toddlers; the most basic task, 1 + 1, can be performed by infants as young as five months and even some members of other animal species. In primary education, students are taught to add numbers in the decimal system, starting with single digits and progressively tackling more difficult problems. Mechanical aids range from the ancient abacus to the modern computer, where research on the most efficient implementations of addition continues to this day."

And you think like .. easy, but then you turn the page:

A software is just a list of bits or we can say binary digits. Next time you developed an application just think you've just found a big "Magic number"! 🤗

When you finally get to join the small group of people that have reached hell.

🤧😢🤧

It's been a long ass journey and it's warm down here.

So I wanted to contact my TV service provider via online chat now in order to do that you need to put in your 9 digits ID number and has you can see in the picture someone thought of a fun way to do it, I'm at 24 wake me when I get to the millions

I have three sons 5 and under. I made them email addresses I intend for them to use as they age the other day. I used their first initial, last name and the last two digits of their years of birth. In binary. Poor kids. They never had a chance.

New developers(5-6 years experience) these days are so pathetic. They dont have any sense of code review. All they want is to put their opinion out without giving any thought.

I had a PR for review today which contains mock specification to match a regular expression and return the corresponding response

The regular expression I put was

104000(02|06|20|48)

Now, this guy comes and puts a comment that we could "simplify" as 104000\d{2}

I replied, the ending digits are not contiguous. The specific pair of digits have to match for these mocks.

Then this guy replied, then we could simplify as 104(0{4}(2|6)l0{3}(20|48)).

I said, I cannot understand how that is simplification. Why do we need such a complex regex to match something very straight forward.

And the guy replied, we should be writing proper regexes, otherwise we could just specify everything explicitly.

I was like WTF man. You try deciphering this next week without taking at least a minute to know which values are matched.

Anyhow, another senior person approved my PR, and I merged it.

So I'm wrapping up for the day and right before I leave a coworker comes up to me with a problem. Our company uses barcodes to track some of our products through their development and we recently switched over to a new system for producing them. The barcodes for this particular product are supposed to have 8 digits, but the last 200 we printed have 9.

I immediately panic because I wrote the script that generates the bar codes and there had been a bug in the past where the script would add extra leading zeroes that weren't supposed to be there. I scramble and check the database, it would be a huge headache if our production database had been compromised with junk barcodes. Nope, all the new barcodes there have the right number of digits.

Next place to check is in the code that writes the barcodes to a text file for staff to print the physical labels from. Nope that's all fine too.

I ask the person who printed out the recent batch of labels to show me how the printing software reads from the text file. She seems confused by my question and shows me how she manually enters in the barcode range to the software. As she does this I watch her add an extra zero to the numbers. 🙃

Even worse there was an option to import all the codes from a text file literally RIGHT BELOW the manual option.

TLDR; Thought my script had screwed up our database, ended up being the fault of a coworker who didn't know how to import text files.

3 in 1
---

If your project needs more than three sentences to describe and the budget is not in the high four digits or preferably five, then you're clearly not fucking paying enough

---

Clients that just write a "?" after you wrote a very detailed tutorial, which even a fucking skunk could do, if given a fucking mouse and keyboard, are just the fucking laboratory perfection made retards

---

I just LOVE IT when my internet drops out for under a minute, so windows doesn't notify you yet and your sftp client trashes both local and remote files

Have fun in the 5-digits club :)

Proud to have been the upvote to give you that last little push :P

Compound infinite list comprehensions in Haskell make my head hurt.

I'll be honest, I've never understood why people say that numbers generated by a computer are pseudorandom and not random.
I know a lot of algorithms for number generation, and I implemented mine, based on time of invokation expressed in nanoseconds, taking digits, manipulating and transforming them. Then I analyzed the probability distribution and it's absolutely flat. So, if you know the Touring test, we can use a modified version of it. If I give you a sequence of random numbers generated by a computer and I give you a sequence of random numbers invented by a person, and you can't notice the difference, so the test is passed.
What's wrong on it?

Just imagine the world where we have CAPITALIZED DIGITS.

As you can see from the screenshot, its working.
The system is actually learning the associations between the digit sequence of semiprime hidden variables and known variables.

Training loss and value loss are super high at the moment and I'm using an absurdly small training set (10k sequence pairs). I'm running on the assumption that there is a very strong correlation between the structures (and that it isn't just all ephemeral).

This initial run is just to see if training an machine learning model is a viable approach.

Won't know for a while. Training loss could get very low (thats a good thing, indicating actual learning), only for it to spike later on, and if it does, I won't know if the sample size is too small, or if I need to do more training, or if the problem is actually intractable.

If or when that happens I'll experiment with different configurations like batch sizes, and more epochs, as well as upping the training set incrementally.

Either case, once the initial model is trained, I need to test it on samples never seen before (products I want to factor) and see if it generates some or all of the digits needed for rapid factorization.

Even partial digits would be a success here.
And I expect to create multiple training sets for each semiprime product and its unknown internal variables versus deriable known variables. The intersections of the sets, and what digits they have in common might be the best shot available for factorizing very large numbers in this approach.

Regardless, once I see that the model works at the small scale, the next step will be to increase the scope of the training data, and begin building out the distributed training platform so I can cut down the training time on a larger model.

I also want to train on random products of very large primes, just for variety and see what happens with that. But everything appears to be working. Working way better than I expected.

The model is running and learning to factorize primes from the set of identities I've been exploring for the last three fucking years.

Feels like things are paying off finally.

Will post updates specifically to this rant as they come. Probably once a day.

Unpaid internships are the worst thing. You exploit young people and promise them experience. Seriously business makes tons of money yet they come with ways to exploit a young person in IT. I think it is evil.

When someone tries to generate getters and setters in C# as if it was Java, you may think: "He is starting!"... And then you ask: "How long have you been working with C#?"

Surprising answer: "1 year" 😱

What have you been doing till now?

I didn't leave, I just got busy working 60 hour weeks in between studying.

I found a new method called matrix decomposition (not the known method of the same name).

Premise is that you break a semiprime down into its component numbers and magnitudes, lets say 697 for example. It becomes 600, 90, and 7.

Then you break each of those down into their prime factorizations (with exponents).
So you get something like

>>> decon(697)
offset: 3, exp: [[Decimal('2'), Decimal('3')], [Decimal('3'), Decimal('1')], [Decimal('5'), Decimal('2')]]
offset: 2, exp: [[Decimal('2'), Decimal('1')], [Decimal('3'), Decimal('2')], [Decimal('5'), Decimal('1')]]
offset: 1, exp: [[Decimal('7'), Decimal('1')]]

And it turns out that in larger numbers there are distinct patterns that act as maps at each offset (or magnitude) of the product, mapping to the respective magnitudes and digits of the factors.

For example I can pretty reliably predict from a product, where the '8's are in its factors.

Apparently theres a whole host of rules like this.
So what I've done is gone an started writing an interpreter with some pseudo-assembly I defined. This has been ongoing for maybe a month, and I've had very little time to work on it in between at my job (which I'm about to be late for here if I don't start getting ready, lol).

Anyway, long and the short of it, the plan is to generate a large data set of primes and their products, and then write a rules engine to generate sets of my custom assembly language, and then fitness test and validate them, winnowing what doesn't work.

The end product should be a function that lets me map from the digits of a product to all the digits of its factors.

It technically already works, like I've printed out a ton of products and eyeballed patterns to derive custom rules, its just not the complete set yet. And instead of spending months or years doing that I'm just gonna finish the system to automatically derive them for me. The rules I found so far have tested out successfully every time, and whether or not the engine finds those will be the test case for if the broader system is viable, but everything looks legit.

I wouldn't have persued this except when I realized the production of semiprimes *must* be non-eularian (long story), it occured to me that there must be rich internal representations mapping products to factors, that we were simply missing.

I'll go into more details in a later post, maybe not today, because I'm working till close tonight (won't be back till 3 am), but after 4 1/2 years the work is bearing fruit.

Also, its good to see you all again. I fucking missed you guys.

A new mathematical constant was discovered recently: Bruce's constant

I took some code from the paper and adapted it in python.

def bruce(n):
J = log(n, 1.333333333333333) / log(n, 2)
K = log(n, 1.333333333333333) / log(n, 3)
return ((J+K)-e)+1

gives e everytime for ((J+K)-bruce)+1, regardless of the value of n.

bruce can always be aproximated with the decimal 4.5, telling you how close n can be used to aproximate e (usually to two digits).

Bruce's constant is equal to 4.5099806905005

It is named after that famous mathematician, bruce lee.

You'll start with four limbs and end up with two in a wheelchair!

I've got at least 4 hours of homework... Now THIS bullshit!!

I got a new debit card from my bank, jumped online, to activate my new card.

I see a picture of my card, with the last 4 digits of my account number show. A big "activate" button right next to it. Sure thing. Click the button, and guess what piece of information I need to verify I am the true owner of the card. Fucking last four digits.

Fucking hell - you just showed me the digits a page ago.

Just a reminder, services like Google Photos are not a backup.

Want to lose nearly three years of photos?
Use Google Photos and do not keep offline back ups of your photos / Takeout archives.

Still on the primenumbers bender.
Had this idea that if there were subtle correlations between a sufficiently large set of identities and the digits of a prime number, the best way to find it would be to automate the search.

And thats just what I did.

I started with trace matrices.
I actually didn't expect much of it. I was hoping I'd at least get lucky with a few chance coincidences.

My first tests failed miserably. Eight percent here, 10% there. "I might as well just pick a number out of a hat!" I thought.

I scaled it way back and asked if it was possible to predict *just* the first digit of either of the prime factors.

That also failed. Prediction rates were low still. Like 0.08-0.15.

So I automated *that*.

After a couple days of on-and-off again semi-automated searching I stumbled on it.

[1144, 827, 326, 1184, -1, -1, -1, -1]

That little sequence is a series of identities representing different values derived from a randomly generated product.

Each slots into a trace matrice. The results of which predict the first digit of one of our factors, with a 83.2% accuracy even after 10k runs, and rising higher with the number of trials.

It's not much, but I was kind of proud of it.

I'm pushing for finding 90%+ now.

Some improvements include using a different sort of operation to generate results. Or logging all results and finding the digit within each result thats *most* likely to predict our targets, across all results. (right now I just take the digit in the ones column, which works but is an arbitrary decision on my part).

Theres also the fact that it's trivial to correctly guess the digit 25% of the time, simply by guessing 1, 3, 7, or 9, because all primes, except for 2, end in one of these four.

I have also yet to find a trace with a specific bias for predicting either the smaller of two unique factors *or* the larger. But I haven't really looked for one either.

I still need to write a generate that takes specific traces, and lets me mutate some of the values, to push them towards certain 'fitness' levels.
This would be useful not just for very high predictions, but to find traces with very *low* predictions.

Why? Because it would actually allow for the *elimination* of possible digits, much like sudoku, from a given place value in a predicted factor.

I don't know if any of this will even end up working past the first digit. But splitting the odds, between the two unique factors of a prime product, and getting 40+% chance of guessing correctly, isn't too bad I think for a total amateur.

Far cry from a couple years ago claiming I broke prime factorization. People still haven't forgiven me for that, lol.

It quite snows digits these days

Wtf. So if I say I'm a web developer and I say I'm from Russia. Then I am automatically a hacker for you? "Web developer + Russian = 95% chance of being a hacker". Yeah, right. Since now, right after I say I'm from Russia I always add this: "No, I'm not a hacker and no, I didn't hack the last election, but I can tell you your last four digits of your SSN if you show me your debit card". Guess what, no one wants to talk to me anymore.

Why? 😪

Addon for converting number digits into words with multi-language support.

This is the most wtf thing that happened me with Javascript, I had a regular expression and it caused bugs only with 4 digits long words, then I just noticed this:

/^.{3}$/.test(null) // false
/^.{4}$/.test(null) // true

What the fuck, I can't believe that who designed the .test method didn't think to avoid null coercion

What follows is an email I want to send very badly right now:

RE: please provide new date format

There is no new date format. You have the day and month switched. In the sample you've provided.

You've been getting away with it up til now because the digits were small enough to be either a month or day.

Then, as soon as you hit a snag you immediately gave up and assumed it was me changing things. You came to me before even making a guess what could be wrong.

Learn to fucking troubleshoot PLEASE. I am sick of your lazy shit coming to me with an issue every time you stub your toe. You are professional developers and I am your junior.

We are going to start accepting credit cards again. Old boss wants to store the tokens in plain text work the last 4 digits of the credit card...

Found a clever little algorithm for computing the product of all primes between n-m without recomputing them.

We'll start with the product of all primes up to some n.

so [2, 2*3, 2*3*5, 2*3*5*,7..] etc

prods = []
i = 0
total = 1
while i < 100:
....total = total*primes[i]
....prods.append(total)
....i = i + 1

Terrible variable names, can't be arsed at the moment.
The result is a list with the values

2, 6, 30, 210, 2310, 30030, etc.

Now assume you have two factors,with indexes i, and j, where j>i

You can calculate the gap between the two corresponding primes easily.
A gap is defined at the product of all primes that fall between the prime indexes i and j.

To calculate the gap between any two primes, merely look up their index, and then do..
prods[j-1]/prods[i]

That is the product of all primes between the J'th prime and the I'th prime

To get the product of all primes *under* i, you can simply look it up like so:

prods[i-1]

Incidentally, finding a number n that is equivalent to (prods[j+i]/prods[j-i]) for any *possible* value of j and i (regardless of whether you precomputed n from the list generator for prods, or simply iterated n=n+1 fashion), is equivalent to finding an algorithm for generating all prime numbers under n.

Hypothetically you could pick a number N out of a hat, thats a thousand digits long, and it happens to be the product of all primes underneath it.

You could then start generating primes by doing
i = 3
while i < N:
....if (N/k)%1 == 0:
........factors.append(N/k)
....i=i+1

The only caveat is that there should be more false solutions as real ones. In otherwords theres no telling if you found a solution N corresponding to some value of (prods[j+i]/prods[j-i]) without testing the primality of *all* values of k under N.

More masturbation with numbers.

If you take some product p,
and do

√p**(1/p)

and it's factors

a**(1/p)

and

b**(1/p)

you might find something interesting.

Take for example

a=21977
b=43331
p=a*b=952285387

(√p)**(1/p) = 1.0000000108551363

a**(1/p) = 1.0000000104986928
b**(1/p) = 1.0000000112115799

More often then not, a, b, or both, will share one or two of the most significant digits in the mantissa, as the root of p.

It doesn't always work, but it seems to be true more often than you might expect.

This is probably obvious in hindsight but I still think it's cool.

In some instances if you then do, say
sqrt(log(p, 1.000000010)), it comes pretty close to the original factors, but thats really hit or miss.

Well I can say one thing for sure. The tooling found for Flutter in the form of IntelliJ and Android Studio is far superior than fucking around on a text editor das for sure.

Not really into the continuous nesting of widgets. But then again I was not a big fan of having jsx either.

Both options still better than fucking around with the Java Android api....and I fucking love Java.

Still feels like Google realized that such approach was better, else I don't think they would have justified the creation of the Flutter lib. And yes. I know that it is not a complete replacement for Android Java, but the interfacing between Dart and the Java api feels more natural at least to me and the widgets look native af so we go on with them big plus digits.

I got pranked. I got pranked good.
My prof at my uni had given us an asigment to do in java for a class.
Easy peasy for me, it was only a formality...
First task was normal but...
The second one included making a random number csv gen with the lenght of at least 10 digits, a class for checking which numbers are a prime or not and a class that will check numbers from that cvs and create a new cvs with only primes in it. I have created the code and only when my fans have taken off like a jet i realised... I fucked up...
In that moment i realised that prime checking might... take a while..

There was a third task but i didnt do it for obvious reasons. He wanted us to download a test set of few text files and make a csv with freq of every word in that test set. The problem was... The test set was a set of 200 literature books...

Can anyone see where this repeats predictably?

Decimal('507.724375') / Decimal('173.2336')

I set the precision to 500+ digits and I'm not seeing any repetitions at a glance.

If there isn't, this is my new favorite irrational number.

Saw lots of regret posts about being in dev field. Then why bother living that way?

Not like engineering, medicine or business management fields, I believe programming needs passion similar to art related careers like acting, music and painting etc. So if you don't have any passion for programming, you won't be successful or satisfy at all.

That doesn't mean it is all good and happy days for every passionate programmers. We sure have ****ed up days (probably more than other fields.) Seriously that's why we have devrants. No? But it doesn't reach to the point of regret to me.

Here our national programming language is probably PHP. The pays are lower than your part-time fast food chain workers. The internet speed is in kbs with 2 digits most of the places. Government doesn't give a crap about IT. No IP copyright laws and so on. I probably would earn more and live better if I were not running this IT business.

But hell yeah I never regret at all.

Guyz let hack this
Here is the tea

A house wife forgot her bank 'ATM PIN' which is a four digit number, but luckily she remembered some hints on how to recall this 'PIN'
Here are some of the clues
(1) The 1st digit is half of the 2nd
(2) The sum of the 2nd and 3rd is 10
(3) The 4th is equal to the 2nd plus 1
(4) The sum of all the digits is 23

What is the ATM PIN ?

Just saw 10000 under my icon and thought of it as a bug in website. With hopes I went on to customize avatar.
2 single digits have never made fun of me that bad.

Not my hack but when I was in university, for one test we had to find all 3 digits number that satisfied current set of criteria. Friend of mine knew from earlier that solution is one number and written program that just prints that number. TA give him 0 points and he get in fight claiming that not only his solution was good but also most optimal one.

So I promised a post after work last night, discussing the new factorization technique.

As before, I use a method called decon() that takes any number, like 697 for example, and first breaks it down into the respective digits and magnitudes.

697 becomes -> 600, 90, and 7.

It then factors *those* to give a decomposition matrix that looks something like the following when printed out:

offset: 3, exp: [[Decimal('2'), Decimal('3')], [Decimal('3'), Decimal('1')], [Decimal('5'), Decimal('2')]]

offset: 2, exp: [[Decimal('2'), Decimal('1')], [Decimal('3'), Decimal('2')], [Decimal('5'), Decimal('1')]]

offset: 1, exp: [[Decimal('7'), Decimal('1')]]

Each entry is a pair of numbers representing a prime base and an exponent.

Now the idea was that, in theory, at each magnitude of a product, we could actually search through the *range* of the product of these exponents.

So for offset three (600) here, we're looking at

2^3 * 3 ^ 1 * 5 ^ 2.

But actually we're searching

2^3 * 3 ^ 1 * 5 ^ 2.
2^3 * 3 ^ 1 * 5 ^ 1
2^3 * 3 ^ 1 * 5 ^ 0
2^3 * 3 ^ 0 * 5 ^ 2.
2^3 * 3 ^ 1 * 5 ^ 1

etc..

On the basis that whatever it generates may be the digits of another magnitude in one of our target product's factors.

And the first optimization or filter we can apply is to notice that assuming our factors pq=n,
and where p <= q, it will always be more efficient to search for the digits of p (because its under n^0.5 or the square root), than the larger factor q.

So by implication we can filter out any product of this exponent search that is greater than the square root of n.

Writing this code was a bit of a headache because I had to deal with potentially very large lists of bases and exponents, so I couldn't just use loops within loops.

Instead I resorted to writing a three state state machine that 'counted down' across these exponents, and it just works.

And now, in practice this doesn't immediately give us anything useful. And I had hoped this would at least give us *upperbounds* to start our search from, for any particular digit of a product's factors at a given magnitude. So the 12 digit (or pick a magnitude out of a hat) of an example product might give us an upperbound on the 2's exponent for that same digit in our lowest factor q of n.

It didn't work out that way. Sometimes there would be 'inversions', where the exponent of a factor on a magnitude of n, would be *lower* than the exponent of that factor on the same digit of q.

But when I started tearing into examples and generating test data I started to see certain patterns emerge, and immediately I found a way to not just pin down these inversions, but get *tight* bounds on the 2's exponents in the corresponding digit for our product's factor itself. It was like the complications I initially saw actually became a means to *tighten* the bounds.

For example, for one particular semiprime n=pq, this was some of the data:

n - offset: 6, exp: [[Decimal('2'), Decimal('5')], [Decimal('5'), Decimal('5')]]

q - offset: 6, exp: [[Decimal('2'), Decimal('6')], [Decimal('3'), Decimal('1')], [Decimal('5'), Decimal('5')]]

It's almost like the base 3 exponent in [n:7] gives away the presence of 3^1 in [q:6], even
though theres no subsequent presence of 3^n in [n:6] itself.

And I found this rule held each time I tested it.

Other rules, not so much, and other rules still would fail in the presence of yet other rules, almost like a giant switchboard.

I immediately realized the implications: rules had precedence, acted predictable when in isolated instances, and changed in specific instances in combination with other rules.

This was ripe for a decision tree generated through random search.

Another product n=pq, with mroe data
q(4)
offset: 4, exp: [[Decimal('2'), Decimal('4')], [Decimal('5'), Decimal('3')]]

n(4)
offset: 4, exp: [[Decimal('2'), Decimal('3')], [Decimal('3'), Decimal('2')], [Decimal('5'), Decimal('3')]]

Suggesting that a nontrivial base 3 exponent (**2 rather than **1) suggests the exponent on the 2 in the relevant
digit of [n], is one less than the same base 2 digital exponent at the same digit on [q]

And so it was clear from the get go that this approach held promise.

From there I discovered a bunch more rules and made some observations.
The bulk of the patterns, regardless of how large the product grows, should be present in the smaller bases (some bound of primes, say the first dozen), because the bulk of exponents for the factorization of any magnitude of a number, overwhelming lean heavily in the lower prime bases.

It was if the entire vulnerability was hiding in plain sight for four+ years, and we'd been approaching factorization all wrong from the beginning, by trying to factor a number, and all its digits at all its magnitudes, all at once, when like addition or multiplication, factorization could be done piecemeal if we knew the patterns to look for.

Sus!
yesterday I bought a cool domain in namecheap, I was very lucky to find short and good one for my case.

Today (at weekends!!!!) I receive a letter:
>Hello **redacted name**,
>
>We are contacting you from the Namecheap Risk Management Team regarding your '**redacted name account**' account.
>
>Unfortunately, your Namecheap account was flagged by our fraud screening system as requiring verification and was locked.
>
>Please follow the instructions below to get your account verified:
>
>- take a color photo of the credit card used for the payment at **redacted link**
>
>Please make sure all of the edges of the credit card are visible, and that we can clearly see the card holder's name, expiration, and last four digits of the card number. The screenshots or images of the card cannot be accepted for verification. >If the submission does not meet these requirements, we can either request to submit the details again or permanently suspend your account.
>
>- provide a valid phone number and the best time to call you (within normal business hours, US Pacific time).
>
>If we do not hear back from you within 24 hours, we will be forced to cancel your orders.
>
>We apologize for any inconvenience that may result from this process. This extra verification is done for your security and to ensure that orders are legitimate. This industry, unfortunately, has a high rate of fraudulent orders, and this sort of >verification helps us drastically reduce fraud and ensure our customers remain secure. Such documents are used for verification only and are not provided to third parties in any way. Account verification is a one-time procedure, after your account >is verified, you will never face this issue again.
>
>Looking forward to your reply.
>
>---------------
>Dmitriy K.
>Risk Management
> Namecheap, Inc.

what if I did not notice it in 24 hours? It is the weekend for god's sake! People usually rest until monday.
They would what, cancel order and scalpel it to super high price?!
I have some doubts if the request is trully having anti fraudulent origins.

What if I used digital visa card? How was I supposed to photo it?

And the service they provided for photoing accepts only photos from web camera. I was lucky that I bought recently web camera with high enough amount of pixel power and manual focus. What if I did not?

That's all really SUS!
The person can not notice the letter within 24 hours time frame until the morning, when it would be already too late.

I'm migrating a site. I moved the host, I need to change DNS servers. It's protected with one time codes sent to an email I don't have access to. The guy who has the access disappeared without a word, it's been over two hours now. I CAN'T DO SHIT WITHOUT THOSE FUCKING DIGITS YOU FUCKER

Past week our Neural Network professor told us after the lecture that the colllege is taking steps for investing in Nvidia Digits DevBox. At first i didnt knw what it is. I googled it and came to knw that it is a fucking beast! Those specs man!!!
Its nice to see that your institute is taking steps to bring latest technology for students.

This kind of BS makes me mad

" - The password must have 6 digits
- It must have at most 2 repeated digits and 3 sequentials"

RIGHT, because 293417 is SO much safer than 999123

Btw, this is a phone company, so with this password you could probably have access to someone's phone number, phone records, address, and much more. WTF

Bootcamps get you up and running in coding quickly. If you are a programmer, companies are only interested on how quickly, error free and cheaply you produce marketable output. Bootcamps enable this.

More or less you are not more than a former assembly line worker putting parts on a car platform. Your value is not very high as you may be exchanged at any time at their will.

Nevertheless, you can earn money quickly. You trade in your youth and time which might be a dead end in the long-term. Trends go to machine learning, artificial intelligence. They will not need Bootcamp people and code workers.

It is better you set up Bootcamps and sell them versus absolving this. Like selling shovels during the gold rush, but not working in the mud of Alaska by yourself.

Your choice is: Making quick money, which fades anyway; or striving for the long-term future proof career.

C/S degrees from Technical Universities of reputation give to you the right direction under a strategic consideration. Companies which pay well, or freelancing with a solid acknowledged background, will always look for top graduates. People from Bootcamps are just OK for hammering assembly line coding. Even worse with SCRUM in one noisy room under enormous team server pressure controls, counting your lines of code per minute, with pale people all around. And groups of controllers never acknowledging nor trusting your work.

To acquire a serious degree, a Bachelor is nothing. Here, in INDIA, Bachelor now is what a former high school grade was. You must carry a diploma or Masters degree combined with internships at big companies with high brand recognition. This will require 4–6 years of your lifetime. You can support this financially by working part-time freelancing as making some projects front- or back-end web, data analysis and else.

Bootcamp people will lose in the long-term. They are the modern cannon fudder of software production.

It is your choice. Personally, I would never do Bootcamps. Quality and sustainability require time, deep studies and devotion.

Is it just me or is the 'storefront' test for CAPTCHA much harder than the schoolbus or traffic light? Could be my diminishing eyesight, or bourbon too.

when your app doesn't show the temperature with two digits

JIRA queue just dropped into the single digits for the first time in weeks. It's getting me right in the feels.

git commit -m “it compiled”
git commit -m “typo”
git commit -m “ugh”
git commit -m “wtf”
git commit -m “ok this doesn’t totally suck”
git commit -m “:shipit:”

So to test my server code that do calculations on large numbers that get above 10 digits 90% of the time.

The end users found a custom web page with basic javascript doing the calculations.

Now I get to explain why that doesn't work.

So first rant, here goes weirdness, and also lengthy rant

So in my company we have the hr and accounting managed by the same person which also deals with all things employee related and she had a need for a way to extract a birthday from, what is in our country the personal identification number, things go great i get a formula that performs parts of the magic up to the point where the first digit of the number dictates the gender and century to be used when forming the full year, mind you only the last two digits of the year are in plain within the id number so i thy a number of ideas. After bashing around google sheets for a while ( i've got open office installed and formulas don't export well to the excel that person uses but google sheets does so i built it there).

First idea : make a few conditionals to check for the value so we have 1 and 2 for 19th century, 3 and 4 for 18th century , 5 and 6 for 20th so i go ahead and write my conditions and they fail, all evaluates to false, it cascades through the else variants up to the last one so i'm wondering if the "if" itself doesn't support the or operator, seems it does, next i think it's the bloody condition written wrong so i reevaluate my logic in php in a test script, it works as intended, then i think ok not the right function called, let's see the docs, docs confirm i'm doing it right but what was wrong was the way i was getting that first number, using left seems to produce a string although the base thing is a number, now i start searching how i can cast it, like you would normaly do when the data type is fried, value function appears to be the solution but it isn't working....now i'm thinking "ok so i have a value and different things to print out so let's look for a switch, maybe it can understand that" switch function found under the form of choice, i get it sorted but am stuck wondering why the heck was the if and value combination not working.

Simple answer to that : value doesn't work well with function results, a known bug listed by someone in a comment, a comment i have failed to read for about 45 minutes of trying to understand.

All in all it worked well for the person asking for it so it's nice.

How do you approach generating "random" unique numbers/strings ? Exactly, when you have to be sure the generated stuff is unique overtime? Eg. as few collisions in future as possible.

Now I don't mean UUIDs but when there is a functionality that needs some length defined, symbol specific and definitely unique data, every time it does it's stuff.

TLDR STORY: Generating 8 digits long numbers so they are (deterministically - wink wink) unique is hard but Format Preserving Encryption saves the day. (for me)

FULL STORY:
I had to deal with both strings and codes today.
One was to generate shortlink word for url, luckily found a library that does exactly this. (Hashids)
BUT generating 8 digits long, somewhat random number was harder then I thought, found out on SO something like "sha256(seed) => bytes => ascii/numbers mangling" but that had a lot of collisions because of how the hash got mangled to actually output numbers and also to fit the length.
After some hours I stumbled upon Format Preserving encryption (pyffx) and man it did what I wanted and it had max 2 collisions in 100k values. Still the solution with this feels hacky af. (encrypting straddled unix timestamp with lots of decimals)

Back again to the horrow show.
We start with the integration. It’s a new project, let’s see how it works. First step: authentication. From the documentation it claims to be an oAuth2. Wait..why just 2 steps to authenticate?! Nevermind, we’ll contact them later. Let’s go on for now.
They need a timestamp with microseconds precision. Here you are!
Nope. Come on! Take the damned timestamp! Nope. Let’s take a look at theirs. If it’s with milliseconds precision, WHY 7 digits after comma?!!!! We decided to contact them. And then.........their answer: we don’t know of any exact number of digits to represent milliseconds.

I see...so it’s arbitrary!!! What are you going to tell us next? One hour can be 3.14159265 minutes then?!!

//long rant ahead!

I need to plan a Wiki with SharePoint for not connected Sites.
Im now in dispute with my CoWorker since 3 Months, this is how the conversation goes. My two bosses are involved in this and also unhappy about SharePoint.

[C refers to CoWorker, M for me]
C: Hey, we finished SharePoint with Selfservice Storage Rooms. They even have a Wiki.

M: Okay cool, will check it out

C: Well we need to also plan the Wiki inside, I already asked our Department Head and he agreed, that you will be the one.

M: Okkkkaaayy, normaly it's your job to do such things, but welp, I will look into it, if we can work with it.

(2 Weeks pass)
M: I checked SharePoint out and tested everything. The Wiki is a Nogo, we need a other solution or programm for ourself a Wiki Integration/Engine. Did you maybe check out Confluence? It has also a SharePoint integration plugin.

C: We wont do Confluence, too expensive (already overspent the budget for SharePoint in six digits 🤬). Also we wont add to SharePoint Custom Code, it needs to stay standard.

M: Thats impossible, SharePoint Wiki is shit and also handels sites just like documents, no brain behind! Also you overspent the Budget and now it's my Problem?!

C: You need to do the best out of it.

(3 weeks passes and we get a meeting with the department heads)

M: Alright I made a UseCase and documented where the essential flaws are in SharePoint Wiki and why we cant use it.

Boss: Ok if it's impossible to use, then we will stay on our Fileserver for Documents and wont use SharePoint.

M: Thats not my Point, my statement is, as status today, SharePoint Wiki is not the right solution, code or buy software to it.

Boss: We will do a Prove of Concept, if it doesnt work then we will aboard it.

M: Well it is only some missing essentials, like hierarchy and Groups for the Pages, Example Confluence has this. If we could built in this features in SharePoint, everything would work out.

C: (angry) I told you that we wont use Confluence!

M: (calm) I said we need Features, not Confluence. Please mind the consent.

(3 weeks passes, and one more meating with bosses)

M: alright here again is a analyses, why already in Theory the current SharePoint Wiki wont work. It's already flawed in the core.

Boss: Yea SharePoint is crap, I checked out confluence and thats a real Wiki.

C: Well I dont know anything about Confluence and never looked at it. But if SharePoint is a fail we need the Proof of Concept.

M: Why do we need to do a Proof of Concept, when it already doesnt work in Theory! Thats nonsence and unlogical.

Next meeting will be in 4 weeks and I will give him the FUCKING PROOF OF CONCEPT. I will be a Bastard and build behind CoWorkers back a Confluence Wiki to show the Departmentheads how to built it right.

I hate CoWorker now, he makes a part of my loved Job a hell, I will goddamn cuk Coworker to space, that fucking Cukatron of lazyness and shit 🤬. I provide the Solutions and you just say no, how dafuq will the project advance, if you always say NO! Are you so unflexible and fixed on your Castle of Ignorancy!

When I said: "version numbers are cheap", I did not expect you to append new digits and identifiers to the existing versioning scheme.

This is the story of probably the least secure CMS ever, at least for the size of it's consumer base. I ran into this many years ago, before I knew anything about how websites work, and the CMS doesn't exist anymore, so I can't really investigate why everything behaved so strangely, but it was strange.

This CMS was a kind of blog platform, except only specially authorised users could view it. It also included hosting. I was helping my friend set it up, and it basically involved sending everybody who was authorized a email with a link to create an account.

The first thing my friend got complaints about was the strange password system. The website had two password boxes, with a limit of (I think) 5 characters each. So when creating a account we recomended people simply insert the first 5 characters in the first box, and the rest in the second. I can not really think of a good explanation for this system, except maybe a shitty way to make sure password are at least 5 characters? Anyway, since this website was insecure the password was emailed to you after the account was created. This is not yet the WTF part.

The CMS forced sidebar with navigation, it also showed the currently logged in users. Except for being unreadable due to a colorful background image, there where many strange behaviors. The sidebar would generally stay even when navigating to external websites. Some internal links would open a second identical sidebar right next to the third. Now, I think that the issue was the main content was in an iframe with the sidebar outside it, but I didn't know about iframe's back then.

So far, we had mostly tested on my friends computer, which was logged in as the blog administrator. At some point, we tried testing with a different account. However, the behavior of sidebars was even stranger now. Now internal links that had previously opened a second, identical sidebar opened a sidebar slightly different from the first: One where the administrator was logged in.

We expirimented somewhat, and found that by clicking links in the second sidebar, we could, with only the login of a random user, change and edit all the settings of the site. Further investigation revealed these urls had a ending like ?user=administrator2J8KZV98YT where administrator was the my friends username. We weren't sure of the exact meaning of the random digits at the end, maybe a hash of the password?

Despite my advice, my friend decided to keep using this CMS. There was also a proper way to do internal links instead of copying the address bar, and he put a warning up not to copy links to on the homepage. Only when the CMS shut down did he finally switch to a system where formatting a link wrong could give anybody admin access.

Itd be cool if we could get something like Schenzen.Io going, but you build the chips from the gates up
Maybe package them into modular units, and connect those at a higher level or abstraction, ad infinitum.

Then add access to virtual LCD output, and other peripherals, or even map output to real hardware, essentially letting you build near bare-metal virtual machines.

Dont know about that last part, but the closest I've seen to the rest is circuit simulator and again, schenzen.

On the machine learning front I figured out I need about ten times as many training samples as validation samples, or vice versa. I'll have to check my notes. Explains why I could get training loss below 2.11

Also, I'm looking at grouping digits, and trying different representations. I'm looking at the hidden variables for primorials to see what that reveals. And I realized because of the amount of configurations and training that I want to do, even a personally built cloud isnt going to be sufficient. I'm gonna have to rent someone else's hardware and run it "in duh cloud."

Any good providers that are ridiculously upfront for beginners to get started with? Namely something cheapish.

analogy for overfitting :
cramming a math problem by heart even the digits of any problem for exam.

now if the exact same problem comes to exam i pass with full marks else if just the digits are changed however the concept is same and simce i mugged up it all rather than understanding it i fail.

Oh god, here comes another math post! I can feel it coming on, like werewolfism during the full moon.

I'm only passingly familiar with logarithms, so this, like everything I've stumbled on, has probably already been discovered, but

n/(1/((n^(1/n))-1))

Is a pretty good approximation (within a couple percentage points, or three or more digits) of the natural logarithm for all the numbers I've checked it on.

For example if

n = 690841693

ln(n) = 20.35342125707679

while our estimate using the above formula comes out to:

n/(1/((n**(1/n))-1)) = 20.353421612948146

Am I missing something obvious here, and if so, what?

Am I doing the idiot savant thing again, or am I just being an idiot again?

I finally got the lstm to a training and validation loss of < 0.05 for predicting the digits of a semiprime's factors.

I used selu activation with lecun normal initialization on a dense decoder, and compiled the model with Adam as the optimizer using mean squared error.

Selu is self-normalizing, meaning it tends to mean 0 and preserves a standard deviation of one, so it eliminates the exploding/vanishing gradient problem. And I can get away with this specifically because selu *only* works on dense layers.

I chose Adam, even though this isn't a spare problem, because Adam excels on noisy problems and non-stationary objectives (definitely this), and because adam typically doesn't require a lot of hyperparameter tuning its ideal here, especially considering because I don't know what the hyperparameters should be to begin with.

I did work out some general guidelines on training quantity vs validation, etc.
The initial set wasn't huge or anything, roughly 110k pairs for training.

It converged pretty quick all things considered, and to the low loss like I mentioned, but even then the system always outputs the same result, regardless of the input, so obviously I'm doing something incorrectly.

The effectiveness of this approach for training and validation makes me question if I haven't got something wildly wrong. Still exploring though and figuring out how to get my answers back out. I'm hoping I just fucked up the output, and not the input as well.

When I arrive 5 months ago to the company, I found that I'll have to work with an internee and someone that has a Masters degree, it is his first job and is in the company for more than a year and a half.
Me thinking, it's gonna be fine, but noooo, it's hell on earth!
What to do with someone like that and he does his work 5x slower and you have to repeat yourself 10 times because he forgets everything?
By the way, the 5x slower when I compare to the internee...

May 2017: They're sunsetting Digits in September, but that's past our runway anyway so we'll worry about it later

September 2017: We have extended the runway this far by letting go of the people who set up our Digits auth and were best equipped to migrate it to Firebase. :|

The amount of digits Python can handle is staggering...

I stopped at 100 000, because I felt that I had proven my point. That's approximately 330 000 bits! (And yes, I had to use Python to be able to calculate that number)

Doesn't work with floats, though...

I'm a duck. A smart duck.

So smart that I know 88 digits of pi.

(a bit late for wk73 but I wanted to post this anyway)

Back in my first year of university, we had to write a relatively simple (though it looked super complicated back then) C++ console application. I don't know what it's called, but it's that game where the computer generates a random 4 digit code and you have to try to guess what it is. Every time you try, it will tell you which digits are correct, which would be correct if they were in a different position and which are outright wrong.

Anyway, the program had a main menu with a help option that would output a short guide on how to play the game. Instead of hard coding it into the source code, the "guide" had go be written in a separate text file and then read and dumped to the screen when necessary.

Here came my great idea on how to read files. Instead of looping through the file until I reached the end, I counted the number of lines my text file had and wrote some gem of a piece of code like this:

for (int i = 0; i<11; i++){
line = file.readline();
cout << line << endl;
}

My teacher obviously took points off for doing such a stupid thing, and I remember complaining A LOT about it. I argued that 11 was a constant because I didn't plan on changing the text file, and that the teacher had no right to take points off for only reading 11 lines because the file only had 11 lines, so it was read in full.

Goddammit, what an innocent little brat I was. I'm glad my first programming teachers were good enough to stay firm and teach me how to do things the right way, even if it's the hard way.

So one of my co-workers had issues with his MKP 2010, it wasn't booting up, wasn't charging, basically acting as a brick. He went everywhere around the office asking other to revive it. I was the last person - my duty was to take out the hard drive and trash it.
I asked him if I can keep it, he allowed me to do so.
After a couple of days I got back to it, installed a hard drive, which I had from my friend from a while ago.
- Played around with charger, got it charging
- Boot it up to recovery mode (I was very excited when I saw that globe spinning)
- The Wi-Fi didn't connect, used LAN
- Formatted drive, installed OS
- Got to endless loop of recovery mode
- Found out that the digits on the keyboard type lower line of letters.. :/
- Boot to safe mode - worked
- Endlessly booted to recovery mode as I tried to get to boot menu
- Realized that the left Shift, Control, and Option does not work
- Used external keyboard and got to boot menu
- Still no success with booting to regular mode
- NVRAM reset!!!!
- Booya! It works!

Should I bother and replace the keyboard? I mean it doesn't matter that much, probably will give it to my mother :)

How can not one, not two, but many many things JUST be so wrong!? like..
Windows. (Yes. THE OS). Why? well... we begin with the garbage, right? the BLOAT.
cortana
mspaint
internet explorer <- wh..WHA?! wh?!?!
ms edge <- okay.. (I saying okay as in a figure of speech I would like to remove it honestly)

why can't I remove internet explorer, and they make another internet explorr called microsoft edge - you guessed it - I can't remove ms edge either.
What's next?! :D
bloat umbruella version -0.1? :D <- a new internet explorer. for 2042.

Cortana. <- some might say "that's not Sooo bad tho". It might not be, but if if it is for me - I would like to remove it.

Okay. okay. moving onto the software.
`V`-Yeah you guessed right. on the first letter.
VISUAL STUDIO.

my face: 😲
I compare visual studio to windows xp to internet explorer to windows overall. they share so much in common...

forced updates,
fixes,
BLOCKS you to compile programs because of NUMEROUS REASONS LIKE..
comment out "CRC303030 whatever" to ignore this message.
you need the build tools vx.x.x.y.x.y.x.t..z.z.z.(100 billion digits later)..x.x.Z OR alternatively you could re-target your solution by (...) (and now today I had enough, I dont see the retarget solution - And I am sure, WHEN I SEE IT - it will just be another problem..

... 💥

I am surprised how windows can run so fluently, with all this crap. Fluently as in actually being running. I am a fan of linux instead though but..
(question to me would probably be why you use windows not linux then?) sometimes I code on windows.. 🤦‍♀️

and it is a pain.

workloads,updates,options,BILLION OF OPTIONS, BILLION OF BUTTONS, stuff I never ever use, takes time to reinstall,install,remove, - windows also needs to restart after each simple thing.... (!?)

sorry. this was nice to write this rant. PHEW! thank GOSH this site exists! 😘 😍

So Fabric has been acquired by Google. Let's pray for Digits, I hope it stays free of charge

@dfox Bug report:

Noticed a problem with the app counting the characters in comments. 3 or 4 times I’ve tried to write a comment that comes down to single digits remaining.

When trying to post I get a message like:

“Oops something went wrong trying to post your comment. It’s probably blank”

But, according to the counter, it should have been fine.

So I realized if done correctly, an autoencoder is really just a bootleg token dictionary.

If we take some input, and pass it through a custom hashfunction that strictly produces hashes with only digits as output, then we can train a network, store the weights and biases, and then train a decoder on top of that.

Using random drop out on the input-output pairs, we can do distillation of the weights and biases to find subgraphs that further condense this embedding.

Why have a token dictionary at all?

When you've managed to get to +99 ok your first day on devRant but can't think of anything else to post to get you into 3 digits!

So I have an array of length 20 which stores a 64 bit decimal number by digits. The starting address is let's say array64. When I am trying to build up my number in ECX:EBX, I am using EDX as the iterator to access the individual elements of the array (I am loading them into AL)

.LOOP:
....
MOV AL, [array64+EDX]
INC EDX
....
JMP .LOOP

The problem is: somehow EDX cannot get higher than 10, so the program just stops and waits for input, when I try to work with a 12 digit number...
This module is outside the main function, and I thought about some Far Pointer problems, but all of my ideas just failed...

What is wrong here?

Posts on Stack Overflow Meta don't effect your rep. I have (randomly) a 140 up vote post on meta which has now awarded me a gold badge. My total rep on plain stack Overflow is still barely triple digits. A couple of bronze and silver badges. I feel somehow conflicted.

I have been trying to use the digits by Twitter Api for web and have been contacting them on everything possible and finding any help for a month and have got almost nowhere. I added ssl and did all the stuff they told me to do.... Wtf Twitter still trying to get this to work

!dev (kinda)

Warning: Might contain (be) stupid rambling.

So I got my new toy and want to play around with it. Just in case I have to return it I first want to make a full disk backup, so I try to boot clonezilla. I press the power button and mash F2, F8, F9 - and it boots straight into the windows setup. Nope, not what I wanted. Try again. And again. Eventually I look it up and apparently I have to hammer the ESC key to get where I want to. Alright, now it works. Boot from USB. Failed. Try again. Failed. Check the BIOS, disable secure boot, reboot. I need to type 4 digits to confirm disabling secure boot. Alright. Reboot, try again, failed. Secure boot is on again. Wtf? After some more infuriating tries I see that NumLock is disabled. AAAARGH. BIOS: Enable NumLock on boot, disable secure boot, enable legacy boot. Input the 4 digits - works! Try to boot from USB: Failed! Grab another USB stick, did the clonezilla image, try again: Finally! It! Works!

Format disk, install Qubes OS. Success!

I have to build a database migration that generates user handles. The user handles are unique within an organization. The user can change them. The auto generated handles are either the first name + last name, or the business name depending on which user type it is. Unless it would be a duplicate. Duplicates auto increment if the handle is taken. The character limit for a user handle is the same length as first name plus last name so I have to check for possible overflow if I add digits. I also have to see if the generated name is in the DB already because a user could have custom entered the result of the auto generation.

This has to be programmed async. The DB driver is using a transaction but multiple calls have to be made to check if the generated handle exists for that organization. Also I have to check the migration script itself for possible duplicates. 3/4 of the users have a handle and with the scale there will definitely be duplicate names.

My idea is if there is a collision, use a UUID and let the users pick something nicer next time they log in. Business says “Reeeeeee!!!! The users shouldn’t see a UUID!!! You can do this!!!” Absurd uniqueness requirements. Absurd backfill procedure. Absurd business rules.

Well fuck Amazon. I am trying to get into my account because for some fucking reason they say my payment method is faulty while they actually write off the subscription of prime of it. But to get into my account I need to login again with 2FA as I have that turned it on. So far so good. But since it's an old phone number I can't login. Well just change the phone number wouldn't you think? Well yes but to change the phone number I need to login in with the old phone number to which I have no longer access 🤦‍♂️. Eventually found a phone number I could call. I get a lovely lady on the phone which guides me to resetting my password but for that, you guessed it, I need to do the 2FA again. I get send through to the next person as she can't change it for me because of privacy reasons (oh well). That guy first askes the last 4 numbers of my creditcard like 5 times because he can't remember it (write it the fuck down then asshole) then he starts mistaking the 6 for 9 (like how the fuck do you do that) and then the text messages don't come in while I am on the phone with him which he tries to blame to my service provider because they would block Amazon (like why would they do that?). But since I got a text message of them 15 min before I shot that down quickly. Then he finally admitted that they might have a disruption going on. So I think we'll fine I'll just ask my question to him how it's possible that Prime stops working as I am watching it because my payment method is faulty according to them (but manage to write off the subscription) and he starts talking just shit. Just admit that you don't know and connect me to someone who does know how that can happen. In the the end I just hung up because I knew I wasn't getting anywhere with this guy and don't you know it, as I start writing this the text messages come in. Problem solved you would say just out that number in the website and you can change your phone number. Well no because I have to tell the number to the guy who I hung up with because the texts weren't coming in 😒. Now I should call them back but I think I'll wait till tomorrow hopefully the day shift will be a bit more knowledgeable on how shit works and can actually remember 4 digits.

Just found out a professor of mine "secures" each student's personal grades by appending a string made up of four random lowercase letters + four random digits (i.e. abcd1234) to the end of a URL.
🙃

More like a confession.

Had a task that involved invalidating a string if it contained more than five digits (anywhere in the string.)

"No problem", I thought. That sounds like a simple regex!

^(\D*\d?\D*){5}$

Turns out catastrophic backtracking had other ideas, and I revert to my usual "atomic grouping fudge that will hopefully work without me really understanding what's going on" of:

^(?>\D*\d?\D*){5}$

...anyone else shamefully follow this mantra, or am I the only one that hasn't skilled up properly on regexes...

Commit message of the day: "Until the [some KPI] can be stored with 15+2 digits in the middle ware, we store its logarithm (to the base of 2) with 3+2 digits."

Trained a hand digit recognition model on MNIST dataset.Got ~97℅ accuracy (wohoo!).Tried predicting my digits,its fucked up! Every ML model's story (?)

Published on BBC, GCHQ have set the challenge below. Would make a fun simple coding challenge. My thought is to brute-force, is there a more efficient way to solve it?

"Take the digits 1,2,3 up to 9 in numerical order and put either a plus sign or a minus sign or neither between the digits to make a sum that adds up to 100. For example, one way of achieving this is: 1 + 2 + 34 - 5 + 67 - 8 + 9 = 100, which uses six plusses and minuses. What is the fewest number of plusses and minuses you need to do this?"

Edit: disclosure: I believe the challenge has passed already and I'm too lazy to enter anyway so don't worry about me or anyone stealing ideas!

!rant

Got a question since I've been working with ancient web technologies for the most part.
How should you handle web request authorization in a React app + Rest API?
Should you create a custom service returning to react app what the user authenticated with a token has access to and create GUI based on that kind of single pre other components response?
Should you just create the react app with components handling the requests and render based on access granted/denied from specific requests?
Or something else altogether? The app will be huge since It's a rewrite off already existing service with 2500 entities and a lot of different access levels and object ownerships. Some pages could easily reach double digits requests if done with per object authorization so I'm not quite sure how to proceed and would prefer not to fuck it up from the get go and everyone on the team has little to no experience with seperated frontend/backend logic.

You could use /\D+/.test('498934') == false to check if a string contains only digits. That statement will result to true. /\D+/.test('oijwei3') == false will result to false since the the test argument has letters in them.

current language vba.
(14 / 24) - (8 / 24) > (6 / 24)
compiled to true. apparently rounding to 8 digits did the trick. quirky was that debug.printing each calculations showed exactly '0.25' for both not giving a hint about some float issue in the first place. ah, and rounding to 4 digits wasn't right either.

I'm beginning to feel like any kind of specific approximation via neural networks is a myth. That if you can't reduce output to simple categorical values that can be broadly interpreted between two points, that it doesn't work.

I have some questions and they don't seem to be getting answered about the design of the net. How many layers should I use ? How many neurons per layer ? How does this relate to the number of desired quantitive scalar outputs I'm looking to create, even if they are normalized, they can vary GREATLY and will if I'm approximating the out of several mathematical expressions. Based on this and the expected error ranges of these numbers and how many possible major digits could be produced within the domain of the variable inputs being introduced, how many neurons per layer ? What does having more layers do ? In pytorch there don't seem to be a lot of layer types per say, but there are a crap ton of activation functions, and should I just be using these at the tail end or should they actually be inserted between layers so the input of the next layer passes through another series of actiavtion functions ? what does this do to the range of output ?

do I need to be a mathematician to do this ?

remembered successes removed quantifiable scalars entirely from output, meaning that I could interpret successful results from ranges of decimal points.

but i've had no success with actual multi variable regression as of yet, even when those input variables are only 2 and on limited value ranges eg [0,100] and [0, 2pi]

and then there are training epochs to avoid overfitting, and reasonable expectation of batches till quality results will start to form.

So for a question on Codeforces, I got the basic logic right, but for one particular test case, the input is a huge number of 250 digits. But the most unsigned long long int can handle is 19 digits. So I used double instead of int, but that makes me lose precision. And I also cannot use the % operator (modulo) which is int only. How do I get around this ?

Brain fart.

In Java and many other languages there are basic types, like char and String. So why does Java have char and String, but not a digit type?

A number is basically a series of digits. For modular arithmetic it is very useful to be able to extract the 3 in the number 1234, it's just the 3rd digit in a number.

Base 2, base 10, base anything could be supported easily too. E.g. a base 2 digit would be:

digit d = 0b2; // or 1b2, but 2b2 would be a compilation error

A number would then be some kind of string of digits.

Any thoughts on this?

I want to print the first number in the Fibonacci sequence to contain over 1000 digits. I got fibbonaci to work but I cannot seem to figure out how to implement the "Contain over 1000 digits".

Do I make a list to store the fibbonaci numbers in then do a statement?

My Python 3 Code:

def fibonacci(num):
if num == 2 or num ==1:
return 1
return sum([fibonacci(num - 2), fibonacci(num - 1)])

print(fibonacci(7))

Let's say AWS assigned you an IP address , which was Pi to 7 digits... would you release it if you didn't need it anymore?