|
Soricidus posted:python v perl: it's not perl the roast beef of plangs
|
# ? Apr 25, 2016 14:29 |
|
|
# ? May 22, 2024 17:36 |
|
gonadic io posted:python is the number 1 lang I recommend to people wanting to learn to code who would be put off by Scratch. I would never ever work in it again myself. and python too but for different reasons
|
# ? Apr 25, 2016 18:28 |
|
i don't really like writing Python for some reason
|
# ? Apr 25, 2016 18:34 |
|
Professional educators say Python is hard to teach on account of the indentation poo poo, which is rather disappointing. I'd totally pick Python as the language for a good top-down introduction to programming, particularly given that Python is used for Real Work(TM) ... insofar as startup poo poo counts as real work, anyway. Like, nobody's impressed when you teach them some obvious mickey mouse poo poo.
|
# ? Apr 25, 2016 19:09 |
|
python is cool if you do a lot with data and run into the limits of what excel can do or handle.
|
# ? Apr 25, 2016 19:17 |
|
there is no limit to what excel can do or handle
|
# ? Apr 25, 2016 19:17 |
redleader posted:what makes python better/worse than any other dynamic lang? i'm somewhat considering going for a python job i've seen advertised Python has a very nice standard library, and a bunch of its third-party libraries are really good too (especially for data science: see matplotlib, numpy, scipy, pandas). Also by plang standards python 3's syntax is pretty nice and has very few surprises (stuff like for-else notwithstanding). Personally, I wish the data scientists had picked a different language to pour all their infrastructure into, but they could have made a much worse decision than python so I guess I should count my blessings.
|
|
# ? Apr 25, 2016 19:25 |
|
Shaggar posted:there is no limit to what excel can do or handle well, except the 2GB limit and also the cannot-open-two-files-with-the-same-name limit
|
# ? Apr 25, 2016 19:26 |
Shaggar posted:there is no limit to what excel can do or handle Well, unless you have more than 1,048,576 data points.
|
|
# ? Apr 25, 2016 19:26 |
|
use 64bit excel instead of 32bit excel.
|
# ? Apr 25, 2016 19:29 |
|
Shaggar posted:use 64bit excel instead of 32bit excel. lmao
|
# ? Apr 25, 2016 19:36 |
|
Wheany posted:well, except the 2GB limit in practice it's more like 1.3gb for some reason
|
# ? Apr 25, 2016 19:37 |
|
teaching python as a first language is surprisingly hard mostly due to stuff that seems like it would make things simpler, such as for-loops being somewhat magical, needing the abstract idea of a sequence, which is actually a huge complication when the students are busy trying to conceptualize the triad of variable-value-type for(int i = 0; i < n; i++) { ... } feels far more arbitrary, but it is unambiguously built out of things the students easily learn anyway (anyone who thinks "i++" is complicated for beginners have all the wrong ideas of education)
|
# ? Apr 25, 2016 19:39 |
|
Shaggar posted:use 64bit excel instead of 32bit excel. seriously though. large orgs have a billion office plugins which require 32 bit office. you can't have 32 and 64 bit office (or office components)installed at the same time, even different versions. most small companies don't even realise they have 32bit office and will claim to be using 64 bit office. 32 bit is still the default installation version of excel in 2016
|
# ? Apr 25, 2016 19:43 |
|
Wheany posted:and also the cannot-open-two-files-with-the-same-name limit
|
# ? Apr 25, 2016 19:59 |
|
Shaggar posted:there is no limit to what excel can do or handle the same is true for using an abacus of arbitrary size
|
# ? Apr 25, 2016 20:00 |
|
I taught python for a while and I'll probably get back into it again soon and to be honest I didn't have many issues with what you guys are talking about. the semantic indentation I hammer it early on and it seems to work. the for loop being dependent on the notion of sequences, well... for people who coded in other languages before I see them getting a bit lazy about trying to take an effort into something that it's seemingly trivial to them, so sometimes I see them trip over it but the hardest thing to teach, by far, is mutation. I try to warn about mutation but man, everybody gets stuck in their own programs because somewhere they modify a list that they were looping over and didn't really notice it.
|
# ? Apr 25, 2016 20:01 |
In college I had a professor write a demonstration-level nuclear reactor simulation in Excel, which on one hand was awesome but on the other hand
|
|
# ? Apr 25, 2016 20:10 |
|
VikingofRock posted:In college I had a professor write a demonstration-level nuclear reactor simulation in Excel, which on one hand was awesome but on the other hand i have some bad news about actual nuclear reactor control systems
|
# ? Apr 25, 2016 20:15 |
|
pointsofdata posted:in practice it's more like 1.3gb for some reason Is the reason that modern Office files (OOXML) are literally zip files with internal package structure, and 1.3gb is about how large a compressed file is which decompresses to something that exhausts a 32-bit address space?
|
# ? Apr 25, 2016 20:22 |
|
VikingofRock posted:In college I had a professor write a demonstration-level nuclear reactor simulation in Excel, which on one hand was awesome but on the other hand excel is pro as hell and you can reference external libs to do lots of stuff.
|
# ? Apr 25, 2016 20:28 |
|
remember when this happened http://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-5-80
|
# ? Apr 25, 2016 20:49 |
|
Internet Janitor posted:Is the reason that modern Office files (OOXML) are literally zip files with internal package structure, and 1.3gb is about how large a compressed file is which decompresses to something that exhausts a 32-bit address space? i dont think they apply any actual compression to the file, they just used zip as a convenient archive format that windows has native support for.
|
# ? Apr 25, 2016 21:00 |
|
VikingofRock posted:Python has a very nice standard library, and a bunch of its third-party libraries are really good too (especially for data science: see matplotlib, numpy, scipy, pandas). Also by plang standards python 3's syntax is pretty nice and has very few surprises (stuff like for-else notwithstanding). whoa now i wouldn't go so far to call mpl good
|
# ? Apr 25, 2016 21:02 |
|
Gazpacho posted:everyone should be put off by scratch, it's the silliest "babby's first" language since PILOT otoh i've successfully taught 8 year old kids using the official scratch tutorials soooo
|
# ? Apr 25, 2016 21:18 |
|
Mr Dog posted:Professional educators say Python is hard to teach on account of the indentation poo poo, which is rather disappointing. imo indentation should be drilled into them from day 1
|
# ? Apr 25, 2016 21:19 |
|
gonadic io posted:otoh i've successfully taught 8 year old kids using the official scratch tutorials soooo scratch is good
|
# ? Apr 25, 2016 21:20 |
|
gonadic io posted:imo indentation should be drilled into them from day 1 lesson 1: ascii art i hope you've all installed my favourite text editor
|
# ? Apr 25, 2016 21:20 |
|
i've given more than one talk on how excel is 1) a programming language and 2) a bad programming language. unless you abuse the h*ck out of named ranges and even then you're in trouble when you need to extend your data model
|
# ? Apr 25, 2016 21:22 |
|
tef posted:lesson 1: ascii art haskell's tab length is 8 spaces, and literally every text editor ever has 4-space default. this messes up so so many people. as their seemingly well indented code is rejected by the compiler
|
# ? Apr 25, 2016 21:23 |
|
gonadic io posted:otoh i've successfully taught 8 year old kids using the official scratch tutorials soooo
|
# ? Apr 25, 2016 21:37 |
|
Gazpacho posted:i assumed we were talking about people children are people, actually.
|
# ? Apr 25, 2016 21:39 |
|
gonadic io posted:i've given more than one talk on how excel is 1) a programming language and 2) a bad programming language. unsurprisingly, it often goes very wrong
|
# ? Apr 25, 2016 21:55 |
|
Symbolic Butt posted:but the hardest thing to teach, by far, is mutation. I try to warn about mutation but man, everybody gets stuck in their own programs because somewhere they modify a list that they were looping over and didn't really notice it. this happened to me when i wrote the first embarrassingly lovely script i ever used for work. open an import file, parse out a list of package names, whack package names against an api endpoint, and hey, if you get a 200, just delete those items and you can use the same file as a list of failed API calls! long story short i think that's just a wall you need to hit and it's probably better to have people work on a problem looping over and modifying a list so they hit that wall sooner rather than later.
|
# ? Apr 25, 2016 21:57 |
|
Suspicious Dish posted:remember when this happened this stuff is the bane of my life. for some reason we treat the formatting of excel cells as significant, and don't use quite the same date formatting as excel or the local user date formats
|
# ? Apr 25, 2016 21:58 |
|
LeftistMuslimObama posted:i dont think they apply any actual compression to the file, they just used zip as a convenient archive format that windows has native support for. pretty sure they do use compression, you're thinking of java jars which typically don't. zips are used in so many formats because they're a standard compression format with support for random access to files (file contents themselves can only be accessed serially), as opposed to say cab files which are indexed but use a single compressed stream of a concatenation of all files, or tar archives which aren't even indexed
|
# ? Apr 25, 2016 22:54 |
|
hackbunny posted:pretty sure they do use compression, you're thinking of java jars which typically don't. are you sure? I'm pretty certain they're compressed by default and I've never heard anyone recommend disabling that before?
|
# ? Apr 25, 2016 23:19 |
|
dragon enthusiast posted:whoa now i wouldn't go so far to call mpl good same, i put it in the "mostly usable" category
|
# ? Apr 26, 2016 00:24 |
|
python hot take: groupby's behavior is dumb
|
# ? Apr 26, 2016 02:08 |
|
|
# ? May 22, 2024 17:36 |
|
hackbunny posted:pretty sure they do use compression, you're thinking of java jars which typically don't. zips are used in so many formats because they're a standard compression format with support for random access to files (file contents themselves can only be accessed serially), as opposed to say cab files which are indexed but use a single compressed stream of a concatenation of all files, or tar archives which aren't even indexed strictly speaking, .zip is an archive format, not a compression specification. "no compression" is legal compression for a .zip file, and each file in the archive doesn't even need to be compressed using the same method. but i'd be astonished if office didn't use deflate like every single other zipper under the sun.
|
# ? Apr 26, 2016 02:36 |