The programmers' humor

[ksju-http]

Английский

It is said that humor in programmer's environment is a specific one. Yes, it often is, because it is mostly based on specific knowledge. Nevertheless (or therefore) I like this kind of jokes very much. I post some of them here, and some - in facebook account. The first ones are usually textual, as this diary is mostly for private, internal use. Jokes posted on facebook are in the form of pictures, because my facebook is mainly a showcase, with attempts to paint pretty, joyful image, that is easy to percept.
However, why not to post these images here too? The attempts to describe what do they depict and why it is funny can be a useful language practice.

So, I wanna be a Machine Learning specialist. But who is a Machine Learning specialist?



Well, the image is clear enough, however, as a newbie specialist I would not say that all data scientists do is writing "from sklearn import svm". That line imports powerful library, that has the most useful machine learning approaches already implemented, but someone can break his/her brains just studding what all these approaches are doing, why they are useful and what kind of tasks they can help to solve. But yes, sklearn, xgboost, tensorflow, pytorch and other tools make things that look incredible from the outer point of view being attainable for a programmer.

Other people think that data scientist's work is like this:



Yes, I often hear some kind of similar information. Having machine learning methods implemented in standard libraries and having learned their correct use, all you should do is to put the data in the needed model. But this «all» turn out not to be so easy! First you have to clearly understand, what should be learned from data, then understand what the available sources of data are, which parts of data in them are useful and which are rubbish, how to transform/clean/normalize the data to make it fit the selected models, how to make huge amount of data being suitable for processing in allowable amount of time, and how to get the most from the small amount of data and so on... And in the end the result has to be evaluated and the process has to be corrected...

And when a data scientists got tired of all this, they go to strike. But, unfortunately, even on the strike they use machine learning terms as slogans:



I'll not explains all these terms separately, but the amusing observation in my opinion is that many English phrases can have two meanings, due to the same word being able to act both as verb and noun. «Support Vector Machine» can be read as «Vector Machine have to be supported!» and as «machine with supporting vectors». Similarly, «Free variable» can be an appeal to free the variables, as well as the mathematical term that designate a variable that is already free

There were a lot of words spoken on possible Machines Revolt and sometimes it seems to be not too unrealistic:



And some companies may even be facilitating them!




Now I'm going to distract from the machine learning topic in order to insert two images that refer to more general programming, and are difficult to embed in the taken narration flow. However, the first one is again about data, precisely, about databases:



It is another word game, this time using the word «table», that can denote a main unit of the relational SQL database and also can be an essential attribute of a pub.
Not all No-SQL databases lack tables, but there are certainly those that do.

That's a funny guy. Possibly I look sometimes like he does.



All the crazy numbers he is telling when being asked about his address, are addresses indeed. They are the IP address of his computer in the network, the local address (IP that a computer can use to refer to itself) and MAC-address (a "physical" address, the unique identifier of a network card).

And finaaaally, one more data-related picture, that was freshly brought to me by facebook. It depicts recent events:



(The guy on the photo is Mark Zuckerberg and the image is joking over the recent scandal about facebook users' private data leakage.)