|
Shaggar posted:you company runs on its reporting and your reporting runs on your data. do not let your application decide how the data works. ever. also invent a time machine and give me 2014 back
|
# ? Apr 2, 2015 21:47 |
|
|
# ? May 13, 2024 15:49 |
|
2015 is much better
|
# ? Apr 2, 2015 21:48 |
|
orms are another thing that are targeted at inexperienced users as something that "just works" when the reality is that they are complicated tools in their own right. the level of experience and skill needed to deploy most orms effectively is high enough to obviate the need for them in the first place. we have smart experienced people here who still stumble on all sorts of poo poo when using poo poo like hibernate.
|
# ? Apr 2, 2015 22:14 |
|
Shaggar posted:you company runs on its reporting and your reporting runs on your data. do not let your application decide how the data works. ever. you have this a little backwards. storing data in a central location allows reporting engines and other applications to access the data without passing through my application. the price of access is that they do not control the schema. it may change without warning, and that's their problem to solve. if they want me to provide a stable API on an ongoing basis, i will happily do that. in my application. and then they can ask my application for the needed data. shaggar has been on the receiving end for so long, parsing data from more important applications, he forgot that it's possible to be the guy writing the critical applications that dictate schema design.
|
# ? Apr 2, 2015 22:32 |
|
Denormalize Yourself And Face To Bloodshed
|
# ? Apr 2, 2015 22:43 |
|
i once liked orms then i used some sql driver thing or whatever and just did poo poo myself it was really easy now i really like sql why did i ever use orms?
|
# ? Apr 2, 2015 22:48 |
|
like now half the poo poo i used to write a bespoke fart app for i can just fire up a db connection and select farts from boners where butts = gay groupby foos
|
# ? Apr 2, 2015 22:49 |
|
Notorious b.s.d. posted:you have this a little backwards. reporting is the only important application and without it you don't get paid and your application is pointless. designing the data to suit the needs of the application instead of the business always results in failure.
|
# ? Apr 2, 2015 23:05 |
|
Shaggar posted:reporting is the only important application and without it you don't get paid and your application is pointless. designing the data to suit the needs of the application instead of the business always results in failure. yeah that's just not true for my job shaggar. i 100% believe it's true for your job, but it's not the case for many others. in a shocking development, one man's experiences in rural maine are not universal!
|
# ? Apr 2, 2015 23:10 |
|
also shaggar is not actually wrong, he's just over-generalizing it very often makes no loving sense for the app to drive the data design. it's really dumb to design ORM frameworks that take it as a given that the app can alter the schema at will sometimes the app drives the schema, sometimes it doesn't. for your orm to be any good, it needs to be able to handle both corner cases and a wide variety of situations in between
|
# ? Apr 2, 2015 23:12 |
|
I kind of get why mongo etc. seems like a good idea because you can go json to page directly using node or whatever the cool kids use now but seeing as .net can serialise a class to json and back with literally zero effort you're pretty much just saving yourself a few lines of casting types from a db query and that's it. otoh i also think stored procs are great which apparently makes me a dinosaur and i am also a terrible programmer so whatever
|
# ? Apr 3, 2015 00:47 |
|
i would like to back up my terrible programmer claim here by adding that i spent about 6 hours trying to add a label to a line in an svg diagram over the last couple of days before realising that i was copying the wrong poo poo off stack overflow(!) and was breaking the svg spec (which i had not bothered to look at). This is not really even my job either. Stored procs are good though.
|
# ? Apr 3, 2015 00:56 |
|
stored procs are great
|
# ? Apr 3, 2015 01:00 |
|
today i spent 45 minutes trying to figure why my commit was trying to upload 190MB file that was in my gitignore i ended up doing a git rebase so no one would notice i'm an idiot
|
# ? Apr 3, 2015 01:33 |
|
Sitting in a two hour meeting to bikeshed organizing team projects in TFS since its slow since it's on an old server with a sql server running on it. Lmao
|
# ? Apr 3, 2015 19:10 |
|
|
# ? Apr 4, 2015 11:24 |
|
it disturbs me that I immediately knew what that was
|
# ? Apr 4, 2015 11:41 |
|
do the hieroglyphics have a special meaning in this?
|
# ? Apr 4, 2015 12:10 |
|
fart simpson posted:do the hieroglyphics have a special meaning in this? no, just unicode variable names
|
# ? Apr 4, 2015 12:24 |
|
missed a good pizza to pileofpoo conversion opportunity
|
# ? Apr 4, 2015 18:24 |
|
MALE SHOEGAZE posted:Luigi is extremely sick can you elaborate on this because I have to use it at work and so far it has mostly been a huge pain in the rear end
|
# ? Apr 4, 2015 18:24 |
|
Corla Plankun posted:can you elaborate on this because I have to use it at work and so far it has mostly been a huge pain in the rear end Well if it's been set up poorly I can see how it would be bad. Like, if your individual tasks aren't divided into discrete units of work then it's going to suck and be pointless. Can you explain how it's causing you problems?
|
# ? Apr 4, 2015 21:01 |
|
Bloody posted:i once liked orms orms looked attractive to me at first because I really didn't like phpmyadmin
|
# ? Apr 5, 2015 10:48 |
|
also + for string concatenation makes me a little uncomfortable because commutativity and stuff lua got it right by using .. instead imo
|
# ? Apr 5, 2015 10:52 |
|
Symbolic Butt posted:also + for string concatenation makes me a little uncomfortable because commutativity and stuff yeah it is the overloading that is offensive. "+" for strings doesn't behave like "+" for numbers i'm not sure that string concatenation is such a common operation that i need an operator for it, but if i did, using something non-math-y seems like a good choice
|
# ? Apr 5, 2015 15:36 |
|
Symbolic Butt posted:also + for string concatenation makes me a little uncomfortable because commutativity and stuff at least + for string concatenation obeys the rule that (a + b) + c == a + (b + c) for all abc, unlike e.g. doing maths with numbers as implemented by eg javascript
|
# ? Apr 5, 2015 15:53 |
|
MALE SHOEGAZE posted:Well if it's been set up poorly I can see how it would be bad. Like, if your individual tasks aren't divided into discrete units of work then it's going to suck and be pointless. the biggest problem so far has been the way it traverses task trees i set up a backfill that was broken up into discrete tasks that seemed sensible but it did the 0th task for each day and then the 1st task for each day, etc, etc, instead of finishing one day at a time which lead to hive getting saturated with a lot of table creation tasks and locking up i dont really understand this kind of big data infrastructure yet so i'm sure i'll figure it out eventually. so far it just seems like having all of this poo poo in java makes everything insanely slow and hard to troubleshoot
|
# ? Apr 5, 2015 16:01 |
|
Notorious b.s.d. posted:i'm not sure that string concatenation is such a common operation that i need an operator for it, but if i did, using something non-math-y seems like a good choice I rarely use concatenation in ~*~real coding~*~ but it's handy for scripts, prototypes, debugging and stuff this was one of the things that I really missed when I first started learning C (if only someone would've told me about using a watch window debugger thing......)
|
# ? Apr 5, 2015 16:06 |
|
+ for string concatenation is fine and only incredibly stupid people have been confused about the effects of it
|
# ? Apr 5, 2015 17:25 |
|
concatenation is a perfectly reasonable additive combination when you're talking about lists of stuff, whether they be lists of tuples, integers, characters, whatever. allowing it for lists of characters but not other lists is a bit weird though. talking about how concatenation isn't commutative is beside the point really, lots of mathematical operations are commutative over the reals but not other domains, you don't see people claiming matrix multiplication is somehow not multiplication.
|
# ? Apr 5, 2015 17:44 |
|
Jabor posted:concatenation is a perfectly reasonable additive combination when you're talking about lists of stuff, whether they be lists of tuples, integers, characters, whatever. allowing it for lists of characters but not other lists is a bit weird though. there's a well established convention (since abelian groups I think) that you usually denote a generic commutative operator by the plus sign. sometimes it's even an implicit thing, you see some plus signs that the author never really cared to define because it's implicit by convention that it's some commutative operation I mean, I know I'm being a pedantic rear end here with arbitrary mathematical jargon, but this is HOW I FEEEEL after wasting my time studying algebra
|
# ? Apr 5, 2015 18:31 |
|
just use ++
|
# ? Apr 5, 2015 18:37 |
|
anything which stops people using ++ for its current purpose I guess
|
# ? Apr 5, 2015 18:40 |
|
anyway serious question: what's the deal with the comma operator in C? I mean what the gently caress
|
# ? Apr 5, 2015 18:44 |
|
Symbolic Butt posted:anyway serious question: what's the deal with the comma operator in C? very, very occasionally it's useful to do something along the lines of: for (int i = 0, j = 10; whatever; ++i, --j) in general though yeah shoot anyone who uses it
|
# ? Apr 5, 2015 19:06 |
|
Corla Plankun posted:the biggest problem so far has been the way it traverses task trees I'm just cargoculting on a bunch of work a coworker did, so I'm not 100% sure whether your breadth-first versus depth-first traversal issue is something that could be easily solved with a configuration setting, but it seems like it should be. And having to do poo poo in java seems like a consequence of using hive/hadoop, not using luigi.
|
# ? Apr 5, 2015 19:58 |
|
Symbolic Butt posted:there's a well established convention (since abelian groups I think) that you usually denote a generic commutative operator by the plus sign. sometimes it's even an implicit thing, you see some plus signs that the author never really cared to define because it's implicit by convention that it's some commutative operation terms and notations in algebra are universal and well-established as long as you carefully only read one textbook. good luck finding a wikipedia page about an algebraic structure that doesn't have a paragraph talking about how different authors use the term for slightly different things i mean, you're right, + is usually commutative, but appealing to the notational consistency of algebraists of all people is still pretty funny
|
# ? Apr 5, 2015 21:38 |
|
why does OpenCV has its own String implementation??? is this standard practice for c++ projects?
|
# ? Apr 5, 2015 22:02 |
|
Yes, apparently STL Strings waste too much memories.
|
# ? Apr 5, 2015 22:08 |
|
|
# ? May 13, 2024 15:49 |
|
Symbolic Butt posted:anyway serious question: what's the deal with the comma operator in C? another thing for the tall pile of features of C which nobody should use
|
# ? Apr 5, 2015 22:31 |