Warning: Attempt to read property "comment_date" on null in /var/www/wptbox/wp-includes/comment-template.php on line 1043
Warning: Attempt to read property "comment_date" on null in /var/www/wptbox/wp-includes/comment-template.php on line 1043
Warning: Attempt to read property "comment_date" on null in /var/www/wptbox/wp-includes/comment-template.php on line 1043
The missus wants to start doing computer stuff for work. I told her to learn python cause i've heard that it's pretty basic and the beginning coder can quickly produce code.
I watched 3b1b and see that he uses Julia. I had a look at the sticky and it says nothing about Julia. Should i get the missus to learn Julia instead?
>Julia is a high-performance, dynamic programming language for modern technical computing. It solves the two language problem by being simultaneously as fast as C and as easy as Python. The built-in package manager provides access to over 6,000 Julia packages created by the Julia community, in addition to foreign function interfaces for tens of thousands of C, Fortran, Python, and R libraries.
It doesn't matter much. Python is much more ubiquitous so it's easier to find guides and help. Also Julia is much younger so it has less packages but it has enough for most things you'd need to do. Julia's package manager is much better than python's because python is pretty dated but probably not important for basic stuff. Either way both are going to be about the same difficulty to learn and it's more about learning something how to code than the specific language.
As far as Julia's claims that it's the greatest language that's ever been written, it's not. They're pretty full of themselves. In my experience Julia never comes anywhere near what C does in speed. I don't know what they base those claims off of but it's clearly a specific case that Julia is highly optimized for.
would potential employers care which language she knows? or do they just want (extreme) competency in a language?
i dig rocks for a living so sorry for daft questions
Oh if she's looking for a new job that has to do with programming, python is way more marketable. I was thinking she just wanted to learn to automate some stuff in her current job.
It entirely depends what qualifications she has and what industry she is in.
Python is becoming the de-facto language for scientists, because it's easy to learn and has ton of numerical libraries.
So if she's a geologist, geneticist, biologist or anything like that, Python is a safe choice.
Generally Python is rarely a bad choice of language to learn.
In any case I would not bother with Julia, it's a fairly niche language.
>would potential employers care which language she knows?
Depends entirely on the job, for example if you were going for a Data Engineering role (usually Python, SQL and maybe some meme like Scala needed) and you only had deep knowledge of Javascript (used for web browsers etc) then that wouldn't be a good fit, but realistically if you were that good at JS you'd be going for different roles anyway
If your gf is female, she will not need to know any programming to get a job. Worry not
>They're pretty full of themselves.
elaborate
Them consistently pushing that Julia is a fast as C. It's not. It's one thing for vectorized code to be about C since that's really just running fortran code but writing things in for-loops does not compile to code nearly as fast as C. They also always push that all their features make it the best language to do anything. I like their meta-programming, it's nice but like it's not really a game changer. It's a solid language that I like but they always talk as if they've written the perfect language.
Stick to one. It's pretty easy to pick up new languages once you good a programming in general but there's a lot to take in when you start out so keep it simple. As far as human languages go, I really only see programs written in english. Seems to be the universal programming language but I also live in America. Could be different elsewhere.
>they always talk as if they've written the perfect language.
I don't really disagree with this straw man. as soon as static compilation is actually supported I'll probably use it for everything.
just use python. it's the least awful thing that everyone uses.
cheers
is it worthwhile learning two+ languages concurrently?
Also, missus is multilingual. should i encourage her to learn it all in english (cause english coders are better) or learn in different languages (cause cultural perspectives, greater understanding, blah blah blah)?
Languages aren't that big or that hard. Definitely learn two or more. Cpp or at least C should be learned because it's still the standard 'real' language. Python is the standard baby language. Html and JS are the standard internet languages and latex is also a language used by scientists just to write stuff. The massiveness of a language comes from preexisting code called libraries which are like blueprints anyone can use. But learning how to code per se is about concepts and different languages don't just look different, they act different. A c program is written and then compiled into an exe file and dll files if it was dynamically linked. A python program isn't really a program. It's like html file. You open it in python and python runs your instructions on the fly. You can even open python and type directly into it and run code. It's very easy to draft programs or algorithms.
I would hope she knows excel but even excel has some pretty powerful features. But yeah languages all have different features. And it's hard to learn what is sometimes, you gotta also learn what it isn't. To really understand what python is, learn a little about C. They are very good languages to contrast. One is very strict the other will literally guess what you're trying to say.
Also even C is declarative like python. Other languages use other paradigms. You don't really tell the computer exactly what to do, you give it facts and functions and it tries to figure out how to do it
>Languages aren't that big or that hard
No, but getting to the point where you're comfortable in the ecosystem/tooling and understanding how to write and debug solid, idiomatic code in the language takes much more time. And you're more likely to reach the point of deep understanding by focusing on one than splitting your attention too much.
Once you know a programming language (any language, including Python) it's more important to dig deeper and understand what the programs are actually doing at lower levels (the data structures and protocols along with decent understand of operating systems and computer architecture). If you're an actual Computer Scientist, learning C is definitely worth it for the OS/architecture part, but if you're just trying to get work done in the real world, it's a mistake to focus too much on additional languages as opposed to technology more directly related to your work. IOW if you are in high-level data science, learning how to write a more efficient Python program is a better use of your time than learning C, even though learning C might indirectly lead to better understanding of how to write memory-efficient Python programs.
For sure there are tons of skills that are transferable but you do have to focus on one language to ever get good at it. But I do think in the initial stages it actually helps. Learning x vs y. How do you know what interpreted is without knowing what compiled is? You get familiar with the general environment and that determines what language you use. Python is definitely the standard, for its environment. But even then there are plenty of interpreted languages. She could at least check out Java Julia Lua etc. As a beginner you may just not get it. If you just don't understand a concept, another language may implement it in a way that makes more sense to you. OOP for example. Then you get over that learning hump. It's not always worth it to learn multiple things, but even specific coding problems have multiple solutions. Something as simple as sorting. Coding is very ad hoc and the difference between a good coder and a bad one is just all the little things, realizing why one solution is better than another
https://github.com/3b1b
his videos are all made with python, what are you talking about? Is this some stealth julia or 3b1b shill?
fuck me, what a dumb cunt
his most recent video linked to his Julia presentation on MIT. thought he used Julia for his videos. cheers
tell her to learn Ada
it's specifically a programming language for w*men
I've worked in data teams and can confidently say absolutely no one uses Julia
The specs on paper don't matter, what matters is do people actually use it in practice. In practice they EXCLUSIVELY use Python.
At a push you might get some teams using R or Scala or something but in reality it's like 99.9% python then 0.1% meme langs
do you do scientific computing? how does this relate to Julia?
I've worked as a data analyst and data engineer (analyst=calculate figures for normies, engineer=data plumber for analysts)
Julia was designed as a language for data/analytics work, taking some of the nice parts of different data langs (e.g. Julia's REPL is an iPython clone), so it's designed specifically for that purpose.
Only problem with that is no one uses it, whereas Python is used for everything and is easier
Thanks. I'll get her to stick to python. Don't think she'll ever become techy enough to need R
She's a trained accountant but doesn't do accounting. General spreadsheet shit atm but wants to explore something else. Guess python allows the most exploration
cheers mate. so they don't expect you to pick up a new language on the fly then?
get used to it mate: aussies taking over the world
cheers
Python with Microsoft Excel is really good, I think Microsoft is tyring to release offiial support next year. In the mean time check out OpenPyxlr.
i'll use OpenPyxlr to tempt her in. legend
she only had a slight desire to do computer stuff, but i told her that everywhere is desperate for women and she perked up (not her nips unfortunately...). do they like non-white women even more than white women?
I'm a software engineer in Australia and most female developers (which are about 5% to 10% of the profession) tend to be immigrants from India, China, Iran, Pakistan, etc.
The odd white woman is going to be either a butch lesbian or a tranny.
Some employers, especially American companies, are big into "diversity" and shit and will try to promote her.
If she's a trained accountant she could try to get a job as a business analyst, which is a bit of a meme job to be honest. Those are jobs for people who are a bit technical but not too much. In some companies they're useless, in others they actually do some work and help the customer.
ah, not as promising as in america
she already gets bonus super just for being a woman. what a life
>I've worked in data teams and can confidently say absolutely no one uses Julia
Can confirm.
I do HPC Ops/DevOps for biotech and it's overwhelmingly Python and R (R is much higher than 0.1% though, it might even be about half, especially since R Shiny). There are a lot of people using tools written in Fortran, C, C++ and Java, but most of them aren't actually developing in those languages and are just invoking them with parameters as part of a data processing pipeline cobbled together with some mix of shell scripts and Python. I've seen only one support request for Julia in the last 5 years and he using both Python and Julia.
The Anaconda distribution seems to be the most popular way to use Python: https://www.anaconda.com/products/distribution, along with JupyterLab as an IDE/notebook builder.
>In any case I would not bother with Julia
Yeah, basically, learn Python as a foundation and then pick other technologies as may be relevant (which might include Julia at some point, but probably won't).
There's no point in using julia now that JAX exists, which is a shame because julia claims to be fast but has shit TPU support. It's only selling point is speed and syntax sugar, but becomes moot when you can make python as fast.
python doesn't have multiple dispatch and the relevant type system though and never will, which is Julia's main selling point aside from speed
pip install multimethod
No one uses Julia except for MIT grad students and professors
irrelevant language
>the missus
cringe
Julia is something you maybe learn as a side thing as a scientist AFTER you've learned python.
Half the tools you'll need don't exist in Julia.
You use julia because..
just use it ok?
he presented an MIT Julia course linked in his most recent video
So what is wrong with it?
why are you trying to give somebody advice about something that you clearly don't know enough about to give advice about?
Honestly, I bought into the hype of Julia, even to recommend it to my lab teammates, but I feel the community doesn't acknowledge enough its weaknesses to be worked upon and prioritized. I built a whole pipeline with it and it does shine in some places tho.
If it wants to compete against those big languages (say Python), you need better standards for general use (such as easier compiling, more stable and diverse libraries, better documentation, etc).
Check this video, start around the 10 mins mark, 100% describes how I feel it: https://www.youtube.com/watch?v=s6pjxCuNGjc
Oh forgot to add, I think it already surpasses, imho, R and Matlab. But for general purpose languages, it has a long road to make.
damn newbies obssessing over le latest framework and le latest sexy tool and le latest language when they should be learning the fundamentals
It's a self-confessed layman asking on behalf of his normalfag girlfriend/wife.
Simple answer, if its to get a job just go with Python. I basically lied on my application and learned everything in the three days it took to do the assessment they sent to me. I've been learning as I go and I'm really good at it now. I look at code I was writing even 6 months ago and it looks so bad now.
Them some dick sucking lips
Couldn't help but notice
he cute
check the job market in your area and then decide. it's relatively easy to switch to different language later if she has any misgivings about one or the other.