Just like K&R: The C Programming Language, I found this fantastic version. Download from the archive.

This is the best available online version of K&R. I was looking for it everywhere and I could not find it, I looked it up in the chat between me and my friend Antar and I found it there. It should be available now on the link above.

Interesting. This videos shows the reason why ancient Roman heroes and athletes had very small genitals. It relates to self-control and being self-possessed. In contrast, in contract, characters with barbarian behavior would have been depicted with giant genitals.

Related: Where to get news?

I want to talk briefly about a problem that I encounter specially when using RSS feeds, which could be noticed if you kept in mind the feature of social media feeds that this article didn’t mention, which is that most of the time social media platforms get you news/feeds that are relevant to your interests, for example if you post so much about football you will not get sports updates from the BBC about basketball, however you will get when you follow BBC’s RSS.

Some RSS providers overstep this issue by providing feeds per key, and the key is a token that represents the user preferences.

However, the majority do not do that and provide you with a single XML source that include all the updates their site has. What if you only want to get updates about Egypt for example? You do not get to do so, but you can search on the feeds you received for entries having the word Egypt. Which is not very convenient, IMHO.

So I built a very simple customizer to solve this problem, it’s called rssc. You can get it from here https://rssc.fly.dev/. It’s very simple and self-explanatory as you will see in that page, so I do not really have to explain it here. Enjoy.

This does not tell how bad the situation is currently, but it shows how bad it was 2-4 years before each indicator (since a Ph.D takes this time to obtain). The number in 2021 says that the situation was fucked in 2018, and so on. Since the situation currently is no better than 2021, I anticipate that I the rest of new indicator will be even worse.

While this not explicitly addressing the current severity of the situation, this analysis elucidates the extent of its gravity between 2 to 4 years prior to each corresponding indicator, given the temporal span typically required to attain a Ph.D. For instance the value attributed to 2021 serves as an indicator that the economical situation was considerably unfavorable in 2018, and this pattern continues retrogressively. Telling from the absence of any possible improvement since 2021 but only worse crisis, I anticipate that the rest of the indicator for the coming year will be in أسفل سافلين.

See my statement before about wisdom here: What is Wisdom?. I found Marcus Aurelius similiar quote:

I also found this “An everyday activity as a treatment for depression: The benefits of expressive writing for people diagnosed with major depressive disorder” Which is interesting too. I’ve been doing journaling from time to time hoping that I will do some data science on what I will be left with of jounraling entries someday, but I started recently to question the effectiveness of that. Also, I think that another kind of jounraling should be inspected, that is the way used to Marcus Aurelius (his prominent book “thoughts” or “contemplation” was actually intended to be a journal rather than a book); that is to admonish and advice self.

Amicable numbers were known to the Pythagoreans, who credited them with many mystical properties.

Amicable numbers are two different natural numbers related in such a way that the sum of the proper divisors of each is equal to the other number. That is, s(a)=b and s(b)=a, where s(n)=σ(n)-n is equal to the sum of positive divisors of n except n itself (see also divisor function).

People used to credit the amicable numbers for love and harmony, lovers used to share amicable numbers together, a famous amicable pair is 220 and 284.

I think a quick way to solve the question above would be by implementing Sieve’s Algorithm to calculate all the divisors of a number.

If n = p1^e1 * p2^e2 * ... * pk^ek, where each p is a prime number, then the
number of factors of n is (e1 + 1)*(e2 + 1)* ... *(ek + 1)

Don’t really get why this work.

His algorithm:

initial_n = n
num_factors = 1;
for (i = 2; i * i <= initial_n; ++i) // for each number i up until the square root of the given number
{
    power = 0; // suppose the power i appears at is 0
    while (n % i == 0) // while we can divide n by i
    {
        n = n / i // divide it, thus ensuring we'll only check prime factors
        ++power // increase the power i appears at
    }
    num_factors = num_factors * (power + 1) // apply the formula
}

if (n > 1) // will happen for example for 14 = 2 * 7
{
    num_factors = num_factors * 2 // n is prime, and its power can only be 1, so multiply the number of factors by 2
}

Currently studying (Thalji et al., 2017). Unfortunately the author did not provide any implementation to her claims (not any that I could find online) I like the approach though. It’s disappointing that she also didn’t care to mention how they ’automatically’ extracted the roots.

Anyway, I think even such a corpus will be actually incompetent to be tested against because it is still limited to the dictionary resource collection. I also question the method that they used to normalize Arabic, “5) Replace hamza‟s forms ‫ء‬ , ‫آ‬, ‫إ‬ , ‫ا‬ ,‫ة‬ with‫أ‬ .”, it seems reckless.

I will investigate these questions on my paper soon; A Comparative Study of Modern Stemming Methods for The Arabic Language (if there is no link when you click on the title, this means it is not published yet).

My proposed corpus will probably follow the following approach: after collecting all possible patterns from a place like (Amil, 1993), it will be much but not too much work to collect manually. Right after that I shall generate a dataset of all possible permutation which will probably be so big. Most of them will not be even words [roots], which will be validated using the lexicons that (Thalji et al., 2017) actually used. I will not have to extract anything or recomputerize/reformat the computerized versions (like Shamela’s) because all what I will need to do is only search for the permutation and check whether it exists or not, if it does then it’s a valid word.

My corpus implementation will be on https://github.com/larrasket/corpus (if it says not found that mean it is not public yet).

Facial beauty prediction is still a relatively nascent domain; most of the time the results are not credible, + the public dataset they used in the study(Rasmussen et al., 2023) is pretty inexhaustive (5500 for 60 raters) with only frontal faces. The dailymail statement is just too far from what is actually proposed in the original paper. I should write soon about how this relates to the propaganda model.

Pretty clever formula indeed. I believe the best way to think about what it does, is thinking about a normalization for the day number (like convert every Gregorian date to a unified number). There are some obstacles to handle here, like: Leap years. A leap year occurs on any year evenly divisible by 4, but not on a century unless it is divisible by 400 (this is for the orbital period is actually 364.2425). The last term handles this issue. Let’s play with it quickly so you can imagine how the values are normalized. Contemplate the following and try it with other values:

for Y in 1992:1:2018
    result = (Y + floor(Y / 4))
    if Y % 400 == 0
        s = "Y (divisible by 400)"
        println("$s = $Y, Output = $result")
    elseif Y % 100 == 0
        s = "Y (divisible by 100)"
        println("$s = $Y, Output = $result")
    elseif Y % 4 == 0
        s = "Y (divisible by 4)"
        println("$s = $Y, Output = $result")
    else
        println("Y = $Y, Output = $result")
    end
end

Y (divisible by 4) = 1992, Output = 2490.0
Y = 1993, Output = 2491.0
Y = 1994, Output = 2492.0
Y = 1995, Output = 2493.0
Y (divisible by 4) = 1996, Output = 2495.0
Y = 1997, Output = 2496.0
Y = 1998, Output = 2497.0
Y = 1999, Output = 2498.0
Y (divisible by 400) = 2000, Output = 2500.0
Y = 2001, Output = 2501.0
Y = 2002, Output = 2502.0
Y = 2003, Output = 2503.0
Y (divisible by 4) = 2004, Output = 2505.0
Y = 2005, Output = 2506.0
Y = 2006, Output = 2507.0
Y = 2007, Output = 2508.0
Y (divisible by 4) = 2008, Output = 2510.0
Y = 2009, Output = 2511.0
Y = 2010, Output = 2512.0
Y = 2011, Output = 2513.0
Y (divisible by 4) = 2012, Output = 2515.0
Y = 2013, Output = 2516.0
Y = 2014, Output = 2517.0
Y = 2015, Output = 2518.0
Y (divisible by 4) = 2016, Output = 2520.0
Y = 2017, Output = 2521.0
Y = 2018, Output = 2522.0

In the original formula:

\(\scriptsize{{h=\left(q+\left\lfloor {\frac {13(m+1)}{5}}\right\rfloor +Y+\left\lfloor {\frac {Y}{4}}\right\rfloor -\left\lfloor {\frac {Y}{100}}\right\rfloor +\left\lfloor {\frac {Y}{400}}\right\rfloor \right){\bmod {7}}}}\)

The 4 is related to the fact we have a leap year every 4 years (remember, a year is actually 364).

The 100 and 400 comes from the observation that:

What is really clever here is the month manipulation:

So the 5 actually comes from: