terrible programmers: my god, it's full of chars

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > terrible programmers: my god, it's full of chars

«‹›1446 »

eschaton: Mar 7, 2007; Don't you just hate when you wind up in a store with people who are in a socioeconomic class that is pretty obviously about two levels lower than your own?

I imagine MUMPS is edited with something a lot like TECO given the era from which it comes

MUMPS is literally the sort of system that kicked off the crusade for structured programming in the late 1960s

# ? Oct 15, 2015 07:49

Adbot: ADBOT LOVES YOU

# ? May 27, 2024 11:09

Workaday Wizard: Oct 23, 2009; by Pragmatica

LeftistMuslimObama posted:

keep posting :allears:

# ? Oct 15, 2015 11:22

Bloody: Mar 3, 2013

christ and here I complain about verilog like it's so bad

# ? Oct 15, 2015 13:38

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

gonadic io posted:

What are your editing tools like?
I don't think I'd be able to deal with scala if I didn't have an ide - go to declaration, show type of variable, and show implicits in scope save a huge amount of time and effort.

And that's not including any of the refactoring tools like renaming variables across all usages in multiple files.

We have an in house team who maintain a full I DE with intellisense, go to declaration, attached debugging, etc.

No refactoring tools, mostly because you can't trust that no one's assuming a variable in the code downstream.

# ? Oct 15, 2015 14:59

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

Alright, here's a fun program I wrote that demonstrates a few "features" of MUMPS. It's the 99 bottles of beer song, printed 99 down to 0.

code:

d n q,Q,x
	s q=+q,x="s Q=+Q+1,x=""f  s q=q+1 q:(q>(99-Q))  s x=""k Q n Q  s x=""""s Q=+Q  s x=""""""""f  q:q=Q  w $$Q(q) w ! s q=q-'Q""""""""""""""""
	x x x x x x x x x x
	q
Q(q) n Q
	s q=q_" "
	s Q=q_"bottles of beer on the wall, bottles of beer. Take one down and pass it around," _q_" bottles of beer on the wall."
	q Q

Here's some of the fun things demonstrated here:

as you can see, commands can be abbreviated to their first (sometimes first two) character. functions, variables, and commands can all share the same identifiers. the compiler figures out what you want based on context. a MUMPS statement must always begin with a command, and another command can't be issued unless there are first two spaces or a carriage return. This is how the language is able to tell when you want a command vs an identifier. It can tell the difference between functions/subroutines and variables because you must always execute a subroutine with the 'DO' command, and you must call functions using the '$$' syntax, which you can see in the program.
The first subroutine is called 'd'. on that line, you can see i've used new ('n') on the variables q,Q, and x. d is also the command for 'do', so you'd actually run this method from the runtime environment by entering 'd d ^D', because the program lives in the ^D routine.
Q(q) is a function. the 'q' operator is equivalent to "return" in C. If you don't end a subroutine or function with a q, it'll happily carry on into the next subroutine in your routine. Sometimes that's actually good convention (say, by making a "cleanup resources" subroutine that you fall into during normal execution, but can call explicitly if you're making changes and something fucks up and crashes). A lot of the time, it's someone being too clever for their own good. You can tell Q is a function because I quit with the value of the Q variable, which you can see in the last line. Nothing about the declaration at the beginning indicates this, you have to look at the exit points. To use a function's value, you just call it using '$$', like "s var=$$func()" or "i ($$func) d sub"
Another annoying note about q: you can't quit with a value inside an indented code block (say, inside a for loop). You can only quit with a value from the outermost scope of the function. This does mean you see a lot more extraneous result variables in MUMPS functions than you might in other languages.
You escape quotes by just typing two of them. That string in the d function was a loving pain in the rear end to get balanced.
Finally, you see the x operator. In line 2 of d, you can see that I'm executing the code inside the x variable a number of times. Each time, that code is a doing a little processing and then setting x to something new. since that x is from the scope of the function d, that change is able to affect the program even though the x'd code is technically run on a new stack level.
You can also see a couple different flavors of for loop inside that string. I'll talk more about how we never use for loops as for loops another time. While loops are forbidden in our code style manual because they're actually worse than faking while loops using for .

The MUMPSorceress fucked around with this message at 16:21 on Oct 7, 2016

# ? Oct 15, 2015 16:00

cinci zoo sniper: Mar 15, 2013

jesus loving christ

# ? Oct 15, 2015 16:17

Shaggar: Apr 26, 2006

LeftistMuslimObama posted:

Alright, here's a fun program I wrote that demonstrates a few "features" of MUMPS. It's the 99 bottles of beer song, printed 99 down to 0.
code:
d n q,Q,x
	s q=+q,x="s Q=+Q+1,x="f  s q=q+1 q:(q>(99-Q))  s x=""k Q n Q  s x=""""s Q=+Q  s x=""""""""f  q:q=Q  w $$Q(q) w ! s q=q-'Q""""""""""""""""
	x x x x x x x x x x
	q
Q(q) n Q
	s q=q_" "
	s Q=q_"bottles of beer on the wall, bottles of beer. Take one down and pass it around," _q_" bottles of beer on the wall."
	q Q
Here's some of the fun things demonstrated here:

as you can see, commands can be abbreviated to their first (sometimes first two) character. functions, variables, and commands can all share the same identifiers. the compiler figures out what you want based on context. a MUMPS statement must always begin with a command, and another command can't be issued unless there are first two spaces or a carriage return. This is how the language is able to tell when you want a command vs an identifier. It can tell the difference between functions/subroutines and variables because you must always execute a subroutine with the 'DO' command, and you must call functions using the '$$' syntax, which you can see in the program.

The first subroutine is called 'd'. on that line, you can see i've used new ('n') on the variables q,Q, and x. d is also the command for 'do', so you'd actually run this method from the runtime environment by entering 'd d ^D', because the program lives in the ^D routine.

Q(q) is a function. the 'q' operator is equivalent to "return" in C. If you don't end a subroutine or function with a q, it'll happily carry on into the next subroutine in your routine. Sometimes that's actually good convention (say, by making a "cleanup resources" subroutine that you fall into during normal execution, but can call explicitly if you're making changes and something fucks up and crashes). A lot of the time, it's someone being too clever for their own good. You can tell Q is a function because I quit with the value of the Q variable, which you can see in the last line. Nothing about the declaration at the beginning indicates this, you have to look at the exit points. To use a function's value, you just call it using '$$', like "s var=$$func()" or "i ($$func) d sub"

Another annoying note about q: you can't quit with a value inside an indented code block (say, inside a for loop). You can only quit with a value from the outermost scope of the function. This does mean you see a lot more extraneous result variables in MUMPS functions than you might in other languages.

You escape quotes by just typing two of them. That string in the d function was a loving pain in the rear end to get balanced.

Finally, you see the x operator. In line 2 of d, you can see that I'm executing the code inside the x variable a number of times. Each time, that code is a doing a little processing and then setting x to something new. since that x is from the scope of the function d, that change is able to affect the program even though the x'd code is technically run on a new stack level.

You can also see a couple different flavors of for loop inside that string. I'll talk more about how we never use for loops as for loops another time. While loops are forbidden in our code style manual because they're actually worse than faking while loops using for .

that is terrible

# ? Oct 15, 2015 16:19

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

Yeah, there's a reason they pay all the developers 6 figures and give us the best health insurance.

Like I said though, the pain is significantly lessened by strict code reviews that prevent people from using most of this dumbass bullshit in real code. Just about the only place you'll see "x" used is in situations where you'd use a function pointer in another language. It's strictly forbidden if there's any other way to accomplish what you want to do, and variable names are required to be descriptive so you're never going to see someone actually naming variables after functions.

anymore.

but boy have I seen wonders in old code. We actually have one function left in our entire codebase that was written by the founder/CEO. We can't remove it because nobody knows what it does. It's 17 characters, nearly each of which is an assumed variable or an operation on an assumed variable, and it shows up in many, many stack traces. The only comment in the file is her last name and the date it was created, and nobody's ever traced a bug back to it so it just hasn't been worth the effort to work out what it's for.

# ? Oct 15, 2015 16:26

cinci zoo sniper: Mar 15, 2013

stay safe mumps ghost

# ? Oct 15, 2015 16:33

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

kalstrams posted:

stay safe mumps ghost

you will live a long and happy life safe from dated ecosystems

# ? Oct 15, 2015 16:34

jony neuemonic: Nov 13, 2009

i hope i never have to write mumps in my life but these posts are really cool and good, thanks obama.

# ? Oct 15, 2015 16:57

Flat Daddy: Dec 3, 2014; by Nyc_Tattoo

i feel like i shouldn't complain about javascript anymore

# ? Oct 15, 2015 17:30

JewKiller 3000: Nov 28, 2006; by Lowtax

LeftistMuslimObama posted:

Maybe with a bit more experience. The 75k job only lasts for a year and then at the end my work is evaluated and I get promoted to the big-boy 6-figure job. I'm the equivalent of a junior dev or whatever you'd call it right now.

ah, very well. it sounds more reasonable then, especially since madison has a pretty low cost of living. if you enjoy your job, and it sounds like you do, then keep it up, because it also sounds like you're good at it! i'm just saying, don't sell yourself short either, because i would hire you based on the contents of the post i'm quoting, and i'm pretty sure others in this thread would too

# ? Oct 15, 2015 17:40

Dessert Rose: May 17, 2004; awoken in control of a lucid deep dream...

LeftistMuslimObama posted:

but boy have I seen wonders in old code. We actually have one function left in our entire codebase that was written by the founder/CEO. We can't remove it because nobody knows what it does. It's 17 characters, nearly each of which is an assumed variable or an operation on an assumed variable, and it shows up in many, many stack traces. The only comment in the file is her last name and the date it was created, and nobody's ever traced a bug back to it so it just hasn't been worth the effort to work out what it's for.

something something server in a closet wall built in front of the door

# ? Oct 15, 2015 17:42

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

JewKiller 3000 posted:

ah, very well. it sounds more reasonable then, especially since madison has a pretty low cost of living. if you enjoy your job, and it sounds like you do, then keep it up, because it also sounds like you're good at it! i'm just saying, don't sell yourself short either, because i would hire you based on the contents of the post i'm quoting, and i'm pretty sure others in this thread would too

I'll keep that in mind. The one big downside to madison is that I just don't drink that much anymore (ironically), and there's loving 0 else to do for a grown adult. I really appreciate the props, though! I have such severe impostor syndrome going on half the time, so it's nice to hear affirmation from someone not in the mumps trenches.

# ? Oct 15, 2015 17:42

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

Dessert Rose posted:

something something server in a closet wall built in front of the door

lol, if only there was an easy way to change the routine's name to FORTUNATO

# ? Oct 15, 2015 17:43

Dessert Rose: May 17, 2004; awoken in control of a lucid deep dream...

serious tho you sound like you rly know what you're doing

tbh one of the signs of a good programmer is the interest you've shown in doing the best you can with the tools you have and making the tools less painful for others to use

# ? Oct 15, 2015 17:44

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

Dessert Rose posted:

serious tho you sound like you rly know what you're doing

tbh one of the signs of a good programmer is the interest you've shown in doing the best you can with the tools you have and making the tools less painful for others to use

I'd say "should a carpenter not know his tools?", but really all I have is hammers, so I say "should a carpenter not know his hammers?"

# ? Oct 15, 2015 17:48

fritz: Jul 26, 2003

LeftistMuslimObama posted:

but boy have I seen wonders in old code. We actually have one function left in our entire codebase that was written by the founder/CEO. We can't remove it because nobody knows what it does. It's 17 characters, nearly each of which is an assumed variable or an operation on an assumed variable, and it shows up in many, many stack traces. The only comment in the file is her last name and the date it was created, and nobody's ever traced a bug back to it so it just hasn't been worth the effort to work out what it's for.

'whenever we ask her what it did, she just smiles and doesn't say anything'

# ? Oct 15, 2015 17:50

fritz: Jul 26, 2003

LeftistMuslimObama posted:

I'd say "should a carpenter not know his tools?", but really all I have is hammers, so I say "should a carpenter not know his hammers?"

# ? Oct 15, 2015 17:50

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

fritz posted:

This is perfect because we actually have cows on-campus. We have a farm that produces a lot of the dairy and veg that our cafeterias use to cook lunch.

Did I forget to mention that our office buildings look like farm buildings, dragons lairs, harry potter castles, and other poo poo like that? It's ridiculous. I'll make an imgur gallery at some point if people are curious.

This was today's menu in cafe #2 (which looks like the harry potter train station):

# ? Oct 15, 2015 17:54

Fanged Lawn Wormy: Jan 4, 2008; SQUEAK! SQUEAK! SQUEAK!

kalstrams posted:

stay safe mumps ghost

# ? Oct 15, 2015 18:04

lord of the files: Sep 4, 2012

kalstrams posted:

jesus loving christ

# ? Oct 15, 2015 18:13

Captain Foo: May 11, 2004; we vibin'
we slidin'
we breathin'
we dyin'

i am very glad i don't have to deal with any of that but am also very glad you are posting about it

# ? Oct 15, 2015 20:35

eschaton: Mar 7, 2007; Don't you just hate when you wind up in a store with people who are in a socioeconomic class that is pretty obviously about two levels lower than your own?

JewKiller 3000 posted:

ah, very well. it sounds more reasonable then, especially since madison has a pretty low cost of living. if you enjoy your job, and it sounds like you do, then keep it up, because it also sounds like you're good at it! i'm just saying, don't sell yourself short either, because i would hire you based on the contents of the post i'm quoting, and i'm pretty sure others in this thread would too

yeah, based on what you've said so far and how you've been able to deal with something as crazypants as MUMPS, you could probably get mainstream work pretty easily that also sticks a 1 in front of your current salary

be sure to keep an eye on the Chicago market: it's a great area, still quite close by (assuming you or your wife have family near Madison), and it has a wide variety of interesting tech work. the downside would be that your commute would be about 45 minutes each way in the Chicago area. doesn't matter where you live and work, it'd be around 45 minutes. it's like a law of nature.

(if you wind up in the Chicago area Mac/iOS scene, I can introduce you to a few people)

# ? Oct 15, 2015 20:43

MononcQc: May 29, 2007

LeftistMuslimObama posted:

Alright, here's a fun program I wrote that demonstrates a few "features" of MUMPS. It's the 99 bottles of beer song, printed 99 down to 0.

[...]

Is there any movement to write readable mumps, like using full-length variables and function names, and skipping lines here and there?

# ? Oct 15, 2015 21:04

eschaton: Mar 7, 2007; Don't you just hate when you wind up in a store with people who are in a socioeconomic class that is pretty obviously about two levels lower than your own?

MononcQc posted:

Is there any movement to write readable mumps, like using full-length variables and function names, and skipping lines here and there?

I showed my coworker who worked at Epic long ago the 99 bottles code, and he chuckled and said he never wrote it like that, he insisted on always spelling out everything

# ? Oct 15, 2015 21:09

gonadic io: Feb 16, 2011; >>=

MononcQc posted:

Is there any movement to write readable mumps, like using full-length variables and function names, and skipping lines here and there?

That's just code golf though, I imagine normal (code reviewed) mumps is much better

# ? Oct 15, 2015 21:11

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

My VM is busy loving up, so have some more MUMPS lessons.

Let's talk loops.

MUMPS doesn't have a WHILE loop. While the most popular implementation, Cach�, has its own language extensions that add lots of curly-brace language stuff (it's seriously like bad Java), they don't exist if you need to be vendor neutral.

Fortunately, the FOR loop does everything a while loop does. Let's talk about the most common ways to loop in MUMPS.

There's the standard for loop:

code:

f var=1:1:100 d
. w "BUTTS",!

This does exactly what you'd think. One kinda nice thing about MUMPS loops is that you can add pre and postconditionals to them. For example (it's helpful to note here, MUMPS comment lines are prefixed with ";" rather than "//")

code:

;This loop quits before the second iteration of the loop
f var=1:1:100 q:var=2  d
. w "BUTTES",!

;This loop quits after the second iteration of the loop
f var=1:1:100 d  q:var=2
. w "BUTTES",!

Those are not good examples of how you'd actually use that, but this is essentially a very primitive version of returning from inside a loop. This is what I was talking about before when I said you see a lot of conditional variables in MUMPS you wouldn't see elsewhere. If my loop was, for example, looping over 100 values in an array searching for a particular value, and I wanted my function to return the array index of the value, it'd be something like this:

code:

findIndex(array,valToFind) n i,found
	f i=1:1:100 d  q:found
	. s:i(array(i)=valToFind) found=1 ;1=true in MUMPS. Hypothetically everything else is false, but in reality it's kind of a crapshoot, so always use 0 for false.
	i found q i
	q 0 ;or whatever you're using as your error return value

However, these are not the most common looping structures in MUMPS. To tell you about that, first we need to take a detour into talking about MUMPS data structures.

In MUMPS, there are only the following ways to store/retrieve data:
1) A variable. Variables can have one value assigned to them and are always strings. When you do math operations on a variable, MUMPS dynamically turns it into a number of the appropriate precision. This can, of course, lead to gross things like "123AAAAA" being turned into 123 when you do a math operation on it. This can be terrible (and led to many missing penny errors in the very early years of our software). There are also some clever use cases for this behavior, but those are kinda proprietary so I can't share

2)Arrays/globals. Arrays and globals exhibit identical behavior except that an array is stored in local memory and a global is stored to disk (and is persistent in the database). You make something a global by prefixing its name with a "^". If you're clever, you'll notice that the routine names I've been posting also start with "^". This is because in MUMPS all code is literally stored in the database and run from there. Yeah, I know. We had to do some really clever hacking to hide the source code when we started selling overseas because it's hard to enforce IP overseas.

So, how do arrays and globals work? They do not behave like a []=style array. I hesitate to even really call them arrays. They are, basically, B+ trees with their nodes (called "subscripts") indexed by any valid string (up to some long character limit I can't remember off the top of my head. It's a few thousand). What does that look like? Let's see:

code:

;Here's the syntax to populate values in global/array
s ^OBAMA("THANKS")="NOT!"
s ^OBAMA("THANKS","OBAMA")="COMMIE!"
s ^OBAMA("THANKS","TED CRUZ")="JUSTICE"
s ^OBAMA="MUSLIM"

;And you can visualize the tree like this (crappy ascii art warning)
                                    ^OBAMA="MUSLIM"
                                              |
                                         THANKS="NOT!"
                                    /                                \
             OBAMA="COMMIE!"              TED CRUZ="JUSTICE"

A global (or array, but I'm just going to say global from now on) can have an unlimited number of subscripts, and each subscript can itself have an unlimited number of subscripts. If you're programming in a MUMPS database, all of your persistent data structures are based on these globals. As you can guess, I'm not going to say much about how we in particular stow our persistent data in these globals, as that's getting into the proprietary system architecture.

Now, one interesting property of these globals is that the subscripts are always stored in ASCII collation order (or special locale-specific collation orders. We utilize alternate collations for Dutch and Danish locales currently). You traverse these subscripts using the $ORDER function. You can use the $ORDER function inside of a for loop to "walk" a subscript level of a global. For example, this code will visit all subscripts of the THANKS subscript:

code:

n sub
f  s sub=$O(^OBAMA("THANKS",sub)) q:sub=""  d
. w ^OBAMA("THANKS",sub),!

;OUTPUT
;"COMMIE!"
;"JUSTICE"

Basically what's happening here is that we're combining the FOR syntax's ability to do an assignment at the beginning of each loop iteration with the preconditional's ability to quit when we hit a desirable stopping condition. In this case we just quit when $O returns "", which means there's no more subscripts to traverse. If we were instead looking for a particular value, our preconditional might quit when the subscript equalled that instead. You can have as many pre and postconditionals as you want, so you can still quit on "" as an error handling case.

There's other stuff you can do with this. The @ operator allows indirection. You can store part of a global's name in a variable, then access it using the indirection operator. The following code has identical output to the last code:

code:

n sub,glo
s glo=$na(^OBAMA("THANKS"))
f  s sub=$O(@GLO@(sub)) q:sub="" d
. w @glo@(sub)

$na, by the way, provides the specified variable's "name". The name is a string representation of the variable guaranteed to work with @. The common use for this is to create functions that are the "providers" of references to your database's globals so that all code that doesn't directly maintain those globals doesn't have bare references. This means you can change your data model later and you should only need to update the functions that are handing out the $na values of your globals.

So, between the natural sorting of globals and the $O function it's pretty painless to create quickly traversable indexes and create well-structured data models.

Of course, you also end up with jerks using globals with a zillion subscript levels as ghetto configuration arrays and they slowly become eye-stabbingly hard to traverse.

Now, if you go back and look at the sorting algorithm I posted before, you'll see $q used. $q is so rarely used that we don't really even bother to teach people about it internally. What $q does is traverse the "physical structure" of the global. This basically means that you loop over every node and leaf in the global that has a value assigned to it. Each call returns the $na of the next path. $q on our example global would return in this order:
^OBAMA
^OBAMA("THANKS")
^OBAMA("THANKS","OBAMA")
^OBAMA("THANKS","TED CRUZ")

As you can imagine, there's very few cases for needing to be able to traverse every arbitrary subscript, as generally you're going to know which subscript contains the data you want. My sorting algorithm takes advantage of this, though to get a natural sort. How does that work? Well, if all of the subscripts of a global are numbers, they get stored in numerical order instead of ASCII order. The algorithm I posted traverses all the values of an array and puts them into a global where each character is its own subscript but runs of numbers are "clumped" into a subscript. This results in a natural sort. Then you "walk" it with $q and just cat everything back together and store those in an array with purely numeric subscripts so the order doesn't change. Fun!

# ? Oct 15, 2015 21:38

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

MononcQc posted:

Is there any movement to write readable mumps, like using full-length variables and function names, and skipping lines here and there?

Yeah. The 99 bottles code was horrible on purpose. I wrote it for fun. Our actual code review processes emphasize routine and variable names that are descriptive.

MUMPS doesn't allow skipping lines, so if you want spacing in your code you have to have empty comments basically. A lot of people are averse to that because back in the day there was a hard linecount limit on routines and comments actually caused performance problems. Both of those issues are resolved, but people are slow to change. There also used to be a hard 31-character limit on "tag^ROUTINE" names, so we used to enforce an 8-character limit on routine names. Also not so anymore, but you have to remember that this language dates back to a time when OSes only supported 8.3 for filenames and it was possible that a process would only have a few kb of memory.

# ? Oct 15, 2015 21:41

Symbolic Butt: Mar 22, 2009; (_!_); Buglord

mumps is amazing

# ? Oct 15, 2015 21:45

Bloody: Mar 3, 2013

LeftistMuslimObama posted:

the mumps trenches

mods?

# ? Oct 15, 2015 21:49

bobbilljim: May 29, 2013; this christmas feels like the very first christmas to me

LeftistMuslimObama posted:

i dont say this lightly in teh pos... but dude... your are posts are pretty drat good :thumbsup:

# ? Oct 15, 2015 22:40

DONT THREAD ON ME: Oct 1, 2002; by Nyc_Tattoo; Floss Finder

I'm doing elixir and it feels too much like ruby for me to trust it very much

# ? Oct 16, 2015 01:09

pepito sanchez: Apr 3, 2004; I'm not mexican

LeftistMuslimObama posted:

really cool posts about crazy language i'll never try to use. thanks. (not making fun)

your posts do inspire me want to try my hand at haskell. since it's another language i'll likely never use in practice, but likely (or just probably?) make me learn a lot about good programing practices in languages i do know just by banging my head against the wall. thankfully a good number of people in this thread and others have shared decent links to math problems and small projects for any new language enthusiast.

# ? Oct 16, 2015 01:22

DONT THREAD ON ME: Oct 1, 2002; by Nyc_Tattoo; Floss Finder

LeftistMuslimObama posted:

I'll keep that in mind. The one big downside to madison is that I just don't drink that much anymore (ironically), and there's loving 0 else to do for a grown adult. I really appreciate the props, though! I have such severe impostor syndrome going on half the time, so it's nice to hear affirmation from someone not in the mumps trenches.

i was going to say the same thing. how long have you been at epic? im assuming not long since you mentioned still sorta being under evaluation.

anyhow rh really likes me and when i get hired i think im gonna tell them that ill only come on full time if they let me work remote

# ? Oct 16, 2015 01:31

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

pepito sanchez posted:

really cool posts about crazy language i'll never try to use. thanks. (not making fun)

your posts do inspire me want to try my hand at haskell. since it's another language i'll likely never use in practice, but likely (or just probably?) make me learn a lot about good programing practices in languages i do know just by banging my head against the wall. thankfully a good number of people in this thread and others have shared decent links to math problems and small projects for any new language enthusiast.

I've never been able to wrap my head around functional languages, but I'm also not a math guy so maybe some of the appeal is lost on me.

I tried to learn prolog briefly after the TA in our Compilers class made a big deal about how cool it was, but ultimately it seemed pretty useless for anything but making toy queries against small datasets. All his demos involved hard coding some statements to query against, and he really floundered if I asked if you could hook it up to a big SQL database or something. And if you've got a big SQL database, chances are you have a more mature platform for querying it than playing with it in prolog anyway.

I think I'm going to start writing a cruddy OS in C or something. That class was fun, and I need to maintain those skills, but I can't contribute to most of the big open source projects because my contract states I can't work for any other business interests and if anyone makes money on my work that counts, meaning stuff like Linux or any of the open source libraries people actually use are out of the question.

Thanks City of Glompton for the glorious sig

# ? Oct 16, 2015 01:34

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

MALE SHOEGAZE posted:

i was going to say the same thing. how long have you been at epic? im assuming not long since you mentioned still sorta being under evaluation.

anyhow rh really likes me and when i get hired i think im gonna tell them that ill only come on full time if they let me work remote

nearly 4 years. Like I said, I started in a different role (which I won't state because it will become super easy to identify me irl then) and have been taking classes at UW in the evenings for the last 2.5 years to pass Epic's requirements for a transition into development roles. This has been a major long-term investment in myself.

In the meantime, I had most of my (much more boring) work automated with various utilities I've written. I handed the keys to all that poo poo to my successor in my old job, so he gets to work 2 hour days for the rest of his life pretty much.

Thanks City of Glompton for the glorious sig

# ? Oct 16, 2015 01:36

echinopsis: Apr 13, 2004; by Fluffdaddy

epic as in ue4?

# ? Oct 16, 2015 01:42

Adbot: ADBOT LOVES YOU

# ? May 27, 2024 11:09

pepito sanchez: Apr 3, 2004; I'm not mexican

i'm not a math guy either. i don't think of solving common math problems in a new programming language as "math guy" challenges. oddly i've always loved math and i've always been terrible at it. i am not an engineering student. but i like thinking of myself as something of an academic, and learning new things related to something you like is always good. it's why i like reading your posts. obviously you're more into the math than i ever was.

personally i just want to start making my own little minimum-graphics game from scratch to learn the most i can (language/structure/algortihm/design-wise) from a personal project. something like youtube's handmadehero, but truly just my own. the problem for me is getting something of an original idea. i've played dwarf fortress about twice, yet i love its idea and what effort goes behind it. maybe when i get more time.

# ? Oct 16, 2015 01:44

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > terrible programmers: my god, it's full of chars

«‹›1446 »