« first day (1092 days earlier)      last day (3892 days later) » 
00:00 - 19:0019:00 - 23:00

enderland
enderland
12:46 AM
@psr yeah that's basically what I figure
 
 
5 hours later…
Robert Harvey
Robert Harvey
5:37 AM
@JimmyHoffa You still around?
 
Matt D
Matt D
whats up?
 
Robert Harvey
Robert Harvey
When you make a method asynchronous in c# 4.0 (using the async keyword), do you really have to make every method that is downstream of the original call also asynchronous?
 
Matt D
Matt D
no idea. I've not used the async primitives in c# yet
check the docs?
 
Robert Harvey
Robert Harvey
Apparently when you mix async code with ordinary code it can cause deadlocks.
...under "Async All the Way"
 
Matt D
Matt D
i'd put your comment under "misleading title"
what they're saying is "dont call async code from sync code"
as it could cause problems if the async code never returns.
its the old RPC problem. async code calls async code which calls sync code which calls original async code which never returns
I've worked on some RPC based systems where you have service A call service B, which calls service A, which does some work and then returns
which is fine if you've got enough instances of service A to service all the calls
suddenly when things get busy. the whole thing deadlocks because service A's are waiting on service B, which is waiting for a free service A
and then you punch the developer who thought that was a good idea
 
Robert Harvey
Robert Harvey
5:49 AM
Wait, what? You're implying that every single method has to be asynchronous, including public static void main
 
Matt D
Matt D
no
 
Robert Harvey
Robert Harvey
Well, public static void main is synchronous, so if you call async code from it...
I suppose the "leaf" methods can be synchronous under your definition.
 
Matt D
Matt D
im implying that you can call async code from sync code. its more what you call from the async code. you can call back into sync code, as long as that doesnt then call into other things.
 
Robert Harvey
Robert Harvey
So you're basically saying it's not reentrant. I could have told you that, and I've never used it (yet).
Sounds like a job for immutability.
 
Matt D
Matt D
well im assuming that :) i've not used it yet either :) i've just used other things
and they say "In particular, it’s usually a bad idea to block on async code by calling Task.Wait or Task.Result."
i mean, at some point you need to start an async call chain
 
Robert Harvey
Robert Harvey
5:52 AM
Because... well, you did want asynchronousness, didn't you? So why block on it?
Doesn't make much sense, really.
 
Matt D
Matt D
people do the darndest things
 
Robert Harvey
Robert Harvey
So I see.
:)
 
Matt D
Matt D
the example they give is pretty good. that'd catch me out.
i wouldnt be expecting such context schenanigans
and the solution they use is to use an async main
interesting
i'll have to read up on this over the weekend
 
Robert Harvey
Robert Harvey
I'll read it as well.
Good night.
 
Matt D
Matt D
gl hf
 
 
2 hours later…
Matt D
Matt D
8:02 AM
lol
 
thorsten müller
thorsten müller
8:44 AM
Anybody knows about some good resources about how to build/design a WEB API? Mainly on the architecture aspects. I will (most likely) build it with RoR.
 
 
3 hours later…
gnat
gnat
11:16 AM
amount and intensity of crap that sometimes comes in can be depressing
-2
Q: How come i cant remember anything in java except JFrame?

user2564491Im a 10 year old programmer. I know over 20+ languages and still remember them. But when i started learning java, I cant remember anything about it except for JFrame? I need help cause i want to be like notch. I know he studied java for a long time but i was learning it for 3 years and STILL only...

java
-4
Q: Do programmers need quiet? Do they get it?

Andyz SmithDo programmers need quiet? If they do why do so many places not have quiet and still seem to get something done? Do programmers get quiet? If no, WHY not. PLEASE, references and substantive answers based on notable sources please! DISPOSITION Note this question was asked in a way specificall...

project-management work-environment job-definition
having two Qs of that kind shown in the row is sad
answers like these don't add fun, either
-1
A: Is there a name for the concept of a "cumulative checksum"?

MSaltersIt's called a hierarchical checksum.

 
thorsten müller
thorsten müller
Well, people are used to forums like this to work like a chat room. Drop in anything and get some more or less relevant discussion. Obviously most of the 'bad' questions come from new users.
2
They expect the common question and answer cycle. Here is my problem, now ask about details, then give help. Obviously we expect a bit more from our users. But that means that we have to take the new ones by the hand and show them around. Sometimes it works, sometimes it fails.
But then, this one is really bad:
-5
Q: need to develop a operating system for mobile phone

user27522I'm planning to start up a new business that related to mobile phone operating systems. Well, a customer of mine wanted me to develop a fresh operating system for his yet to launch smartphone. Could you please advise me is it advisable to start a linux based operating system in that phone like ub...

operating-systems
 
enderland
enderland
11:56 AM
@thorstenmüller you should stop by The Workplace sometime, we get crap all the time
 
thorsten müller
thorsten müller
@enderland I think one reason is that we recommend it to people whose questions we close here :)
 
enderland
enderland
@thorstenmüller hmmmmm maybe I'll start throwing the "want to be coder my life sucks" questions back over here...
 
thorsten müller
thorsten müller
@enderland They normally ask here first.
It's the "it has buttons lets ask the IT guys" phenomenon.
 
enderland
enderland
lol someone asked a q in chat and was directed to you guys this morning
in The Water Cooler, 3 hours ago, by Vame
Hi folks,

I would like to setup a new workstation for backend development and would be glad to hear some opinions on the device selection.

I currently work with a Macbook pro retina 15". The development will be mostly for linux systems and maybe some mobile.

The software I use is mainly terminal, vim for python development, eclipse and maybe some XCode. Also, automated scripts, remote server connections etc.

Do you believe one or two thunderbolt displays will increase my productivity? Any suggestions for external mouse? I really like the scrolling and gestures on the magic mouse, but it…
 
thorsten müller
thorsten müller
He is at four close votes by now. Will be closed soon. But it's difficult to decide what's on topic here and what not. We just try to keep this from becoming Stack Overflow's chat room. (Though as I told him he is quite welcome in The Whiteboard chat room)
Just looked, and RhysW actually told him to ask in chat, not to post a question. He got this one wrong. (So RhysW was totally right)
 
enderland
enderland
12:59 PM
@thorstenmüller yeah quite a few questions like that are good in chat
I wish I had a macbook pro retina to work with lol
 
thorsten müller
thorsten müller
@enderland you must work as a developer in a small company with no special IT manager. You basically get everything you claim is necessary for the job.
 
enderland
enderland
@thorstenmüller yeah. i'm in an engineering environment as effectively a solo developer - but at a large company unfortunatley
thoguh I did acquire 2x 24" monitors?
the guy before me at this desk had them :|
I probably shouldn't hang out here too much tbh. be too depressing
is there a tag here reflecting individual developers in non-software dev companies?
 
thorsten müller
thorsten müller
1:16 PM
@enderland no, no special tag. All developers are created equal.
 
Ampt
Ampt
1:29 PM
We're all individual developers at non-software dev companies at heart @enderland
and just don't listen to @JimmyHoffa or @MichaelT too much and you wont be so depressed
 
enderland
enderland
@Ampt :) I have such a weird bacground, undergrad in Mechanical Engineering (yuck) but ended up doing C++ dev/architecture work for 2.5 years in grad school
 
gnat
gnat
@enderland tag for individual devs is solo-development programmers.stackexchange.com/questions/tagged/solo-development - whether they are in SW / non-SW, there is no tag for that and I think rightly so
 
Ampt
Ampt
@enderland my Uni advisor has his Bachelors, MS, and PhD all in Mechanical engineering, and now he teaches Software Engineering full time. Got into it by programming robotic arms as an ME
 
enderland
enderland
@gnat yeah, this is perfect!
@Ampt in undergrad I did all sorts of VBA work since, well, interning with lots of tedious excel work screams "program please"
then I started doing some C++ stuff for my grad research...
@Ampt coding is just lots of fun. Plus I REALLY enjoy architecutre/UX stuff
 
Ampt
Ampt
@enderland you ever been to UX.SE? Some really, really interesting knowledge about UX over there
 
enderland
enderland
1:41 PM
@Ampt yup! my masters was in HCI (and mech e, but that's because, well, it was trivially easy to get an ME comajor ha)
 
Ampt
Ampt
@enderland is there anything you don't do?
 
enderland
enderland
@Ampt lol. I don't do CAD work, and won't, since I find it tedious and boring
@gnat some of these questions are really valuable to read, thanks!
 
user55340
2:07 PM
@Ampt We're realists... and I'm actually a dev at a software consulting company... I was pleased beyond belief yesterday when I came to the realization that everyone at the company from support and HR to CEO understands the Traveling Salesman Problem at some level.
 
Matt D
Matt D
is that because they're traveling salesmen?
or are they trying to prove p=np?
;)
 
user55340
The core company product is about optimizing delivery routing.
 
enderland
enderland
@MichaelT do you have any job openings?! ha
 
user55340
Say you've got 4 stops on your route to pick up things... and after #1, the dispatcher adds another stop to the route. Redo the route on the fly and update the ETA for everyone on the route.
 
Matt D
Matt D
classic P problem
or is it NP
i always forget after a beer or two
 
user55340
2:11 PM
NP
 
Matt D
Matt D
anyway. thats awesome. i guess weighted digraphs are your thing
 
user55340
And doing the TSP for small values isn't that bad.
 
Matt D
Matt D
at least you can always brute force a response
 
Ampt
Ampt
come on... come on
just a little more
 
user55340
The estimation of ETA is part of the key value of the product. You're waiting for the delivery - when will it get there? The vehicle is on stop #1, you're stop #3.
 
enderland
enderland
2:13 PM
so you're saying you don't supply mediacom/quest with this service? who are always wrong about this? ha
 
Matt D
Matt D
that constrains the problem. as you cant juggle the delivery order
 
Ampt
Ampt
anyone doing the review queue?
 
user55340
@Ampt There's one...
 
Ampt
Ampt
YES
1337 rep
my job here is done
 
user55340
(have to find something to downvote now...)
 
Ampt
Ampt
2:15 PM
NO NO NO
 
Matt D
Matt D
lol
 
Ampt
Ampt
I've been planning this for a week hahaha
kljafeliaewnvf
 
user55340
You better get a screenshot quickly...
 
Ampt
Ampt
@MichaelT How could you!?!?!?!?
 
enderland
enderland
1,347
 
user55340
2:16 PM
Its easily rectifiable.
 
user55340
Just find 10x crap answers to downvote.
 
Ampt
Ampt
answers only count for -1?
good to know
 
enderland
enderland
@Ampt you pay 1 rep to downvote answers, DVing questions is free though
 
user55340
You, downvoting a non-cw answer gives you -1.
 
user41796
downvoting candidate:
 
user41796
2:18 PM
2
Q: What are the biggest misfeatures or things unfortunately left undefined in "original" C in hindsight?

hydeFrom Wikipedia: The initial development of C occurred at AT&T Bell Labs between 1969 and 1973;2 according to Ritchie, the most creative period occurred in 1972. If you could write a letter to Dennis Ritchie and Ken Thompson and send it back in time to year 1972, to tell them what they s...

programming-languages c history
 
Ampt
Ampt
 
user41796
questions like that one set me off. Such a blatant poll. Initial answers are crap. IMO, it deserves an instant close. Not sure if it should be deleted or just locked to prevent additional comments and voting.
 
user55340
@GlenH7 Just one? Glen... you need to go for the ones that will do something - data.stackexchange.com/programmers/query/133010/cliff-questions
 
user55340
(note - not all of those are bad questions / answers... just ones that are on the verge with a little shove)
 
user41796
@MichaelT caught my ire since it was near the top of the new questions list
 
user55340
2:23 PM
For example...
 
user55340
-5
Q: Help in writing Agile user stories for Healthcare industry

user96248Could someone provide an example an Agile user story/narrative in a healthcare setting ?. I would honestly appreciate the help as I am interviewing for a BA position. Thanks.

agile
 
Ampt
Ampt
just gotta find those answers that are already heavily downvoted and throw mine on top ;)
 
user41796
@MichaelT Thanks for the delete vote suggestion
 
user41796
@RobertHarvey from my experience, you have to make all of the downstream functions that depend upon the result of that async call be asynchronous. It requires quite a bit of a shift in setting up functions and service methods. That example in "async all the way" is a bit contrived, IMO. There's some willful distortion in that example in order to prove a point.
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey With the async/await implementation there's a few important things that need to be understood, first: An "async" method does not execute, rather it is a definition of a method that can be executed, and when you call it; it returns that method, that's what the "task" is. So when you call an asynchronous method you are simply retrieving the definition for a computation to be run. Now, with that you have to understand there are two possible activites:
One activity is that you execute that task, if you do this you will be kicking it off and then moving on to do something else because it is asynchronous. The other activity is that you wait for the task to complete, now here's where it get's confusing: Because of the way the async/await is implemented it dictates that if your method will await the end of the asynchronous task, to ensure your method doesn't block calling methods from doing useful work, you must make your method async
 
user55340
2:33 PM
@GlenH7 Doesn't even need a delete vote. One downvote will kick it into community automatic cleanup.
 
Jimmy Hoffa
Jimmy Hoffa
What that means is that when you await an async task, you therefore make your method async, at which point your method no longer executes but back to where we started: returns a definition of a computation which can be executed or awaited.
 
user41796
@MichaelT gonna help it on its way....
 
user41796
@JimmyHoffa that was an awesome explanation
 
Jimmy Hoffa
Jimmy Hoffa
This is where people bleed asynchrony up there stack, they don't understand that an asynchronous method should not be awaited, rather asynchronous operations should be fire-and-forget, and they should side effect the resulting action that is desired rather than you waiting for them to finish so you can complete the result. This misunderstanding causes people to await everything that's async.
 
user41796
.NET 4.5 isn't going to help with that problem at all then. Previously, they made it hard for you to wait on the async return. I can see how await is going to be abused
 
Ampt
Ampt
2:37 PM
lol people are waiting on async methods?
 
Jimmy Hoffa
Jimmy Hoffa
The actor model gives a fantastic understanding of how to approach asynchrony in a non-blocking fashion which is quite beautiful, but relies on concurrency which you may not want to meddle with, in which case Continuation Passing Style is an appropriate approach
I strongly suggest anybody dealing with any asynchrony study and understand both of those things very well
 
gnat
gnat
@hyde as long as you can't define "biggest" and "unfortunately" so that that your understanding will be shared by vast majority of readers / answerers (and I think you can't), your question is doomed to receive items or ideas or opinions instead of answers. Hey, it's already happening, can't you see: "I'm sad", "please change", stuff like that — gnat 40 secs ago
 
Ampt
Ampt
our whole system is one big asynch method. Data changes, onChange events fire, and things happen
 
user41796
@gnat where's a mod when you need one? :-)
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt Yeah, anytime you're dealing with hardware this is how it has to happen. Same with the phone system I'm working on, people press buttons, events fire, I queue a message saying what they pressed for another thread to process and return to the hardware observation loop.
 
Ampt
Ampt
2:42 PM
its kind of hard to wrap your mind around at first but it's not all that difficult to understand
 
Jimmy Hoffa
Jimmy Hoffa
On a side note, some of you may have recognized my explanation up there of the async makes very clear: the async/await stuff is a monad. When you return a task, and your caller awaits you, it returns a task which is composed with your task (which may be composed with the tasks of asynchronous methods you call), this goes back to the 2 buckets of apples turned into one bucket of apples with everything from both buckets of apples.
async = return, await = bind
@psr and @thorstenmüller you may find this worth grokking
 
user41796
@JimmyHoffa I recognized the parallels, yes. My temporal, stubborn refusal to prioritize learning FP was the only thing holding me back from commenting in that vein.
 
gnat
gnat
@GlenH7 mod is here, one of the posts I flagged says "...World Engineer♦ 42 secs ago" :)
 
user41796
happy dance
 
Ampt
Ampt
If it common to have multiple classes defined in one python file or should I be breaking these out into their own files
 
user41796
2:51 PM
Not a python person, but that's a pretty common practice across OO languages
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt I think that is fairly common python, I think I would be anti-pythonic on that personally, but it's probably a half and half. Ask on P.SE
 
user41796
not necessarily a good idea, IMO
 
Ampt
Ampt
well they aren't completely unrelated classes. It's usually an abstract base class and 1-2 concrete classes that implement it
probably < 200 LOC total
 
user41796
Don't know it's worth the risk / cost of refactoring, but future dev could diverge from there. The core issue is the baggage you get from the bundling
 
user41796
meh, leave it be
 
Ampt
Ampt
2:53 PM
I'll make it a question for the rep. er....i mean for the knowledge!
 
user41796
if you knew you would have many concrete implementations, then there is an argument for separating them out.
 
user41796
no shame in rep whoring
3
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt If you're working in async, learn what continuation passing style and the actor model are. Knowing these two things well will give you a huge leg up on a huge portion of folks in the no-longer-future of software development which is to be far more distributed and concurrent.
@Ampt I don't tend to like that just because down the road you end up adding one more, then another and another, and because it started all in one they all get appended in one, before you know it you have 6 implementations in a 3k LOC file. Not to mention combining independent implementations means you're far more likely to be working in the file at the same time as another person and then have to deal with source control merges more.
but I don't know how easy/tricky it is in python to link files together, or what other possible reasons people use to explain why python is often all combined in one file.
 
Ampt
Ampt
0
Q: Is it considered Pythonic to have multiple classes defined in the same file?

AmptIn working with python for the first time, I've found that I end up writing multiple classes in the same file, which is opposed to other languages like Java and C++, who use one file per class. Usually, these classes are made up of 1 abstract base class, with 1-2 concrete implementations who's u...

python code-quality code-smell
let me know if I totally goofed anywhere
or if you just generally hate it
woah.. the SE code highlighter is really smart. It realizes that I defined logger as a class and then highlights it later in the code
that or it thinks logger is a keyword
 
Jimmy Hoffa
Jimmy Hoffa
3:09 PM
@Ampt They use google's code prettify (like most) code.google.com/p/google-code-prettify
 
user55340
@Ampt It identifies the language through the tags on the question... otherwise you can hint it.
 
Ampt
Ampt
@MichaelT no, it recognized a class that I had defined earlier in the file and highlighted it as a class
 
jozefg
jozefg
So Scala apparently has macros.. Ick
 
user55340
158
A: Syntax highlighting language hints

Jeff AtwoodNote that this question is a bit obsolete, because we now infer prettify language type based on the tags. See more: Changes to syntax highlighting This is now implemented. In addition to tag inference (a recent change), you can manually specify the language as a hint to Google Code Prettify. T...

 
Jimmy Hoffa
Jimmy Hoffa
@jozefg Scala is a bloody weird language
 
user55340
3:10 PM
Scala is going to be the php of the jvm (ohh! we want that too! lets add it!)
 
Jimmy Hoffa
Jimmy Hoffa
After looking into it a bit, I see some light in it, but still; I think a large part of me would be creaking "It's not haskell!" because it clearly wants to be in many ways, but then it also has a hand full of just strange in it...
 
jozefg
jozefg
It's a good fusion of OOP and functional, but I don't want a fusion of OOP and functional so..
 
Jimmy Hoffa
Jimmy Hoffa
Yeah, I'm still not convinced "multiparadigm" is so good, everyone says so from the standpoint of being OOP folks, but I rarely see anyone who does FP say it, which leads me to believe OOP folks see having FP in their language is great, FP folks don't tend to feel the same about having OOP in their language though.
 
jozefg
jozefg
I certainly don't haha.. Though the idea of having encapsulation enforced in a more finegrain way is sometimes appealing, then I write a typeclass and feel better
Nope frankly I just want first class modules, not objects
 
Jimmy Hoffa
Jimmy Hoffa
@jozefg Yeah and if you really wanted to have encapsulation; you could create a data type with record syntax and create a value that has the methods filled in on it that you want as if you implemented them inside it..
 
user55340
3:16 PM
@Ampt Thats going to take a lot of downvoting to get you back to a 1337 rep. Maybe you should try for 31337 instead?
 
Ampt
Ampt
@MichaelT I've given up. jerk.
of course I could donate some rep...
 
jozefg
jozefg
@Ampt I'll take it :D
 
Ampt
Ampt
any other tags you could think of for that question guys? got Python, Code-Quality and Code-Smell
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT xml element value as in <elem>value</elem> can or cannot have a trailing space in it in valid xml?
 
Ampt
Ampt
@JimmyHoffa heres a chance to spout everything you know about functional programming:
0
Q: F# History Information

Shelby115I'm not certain this question is entirely appropriate for the site, but I tried my best to pick the right SE site. For my current class, I have to pick and research a Programming Language (or popular scripting language) and answer quite a few questions about it, write a paper, and do some exampl...

history f#
 
jozefg
jozefg
3:20 PM
Because higher kinded polymorphism and CLI play badly
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt Eh... I wish I cared about F# more... also that's kind of a bad question
 
Ampt
Ampt
I was thinking you could just go into why functional programming languages are great and all that
because F# was just developed as .NETs functional language of choice right?
 
jozefg
jozefg
@Ampt Yes, historically, it was also because Haskell couldn't be moved to CLI. (MSR employs the two GHC gurus)
 
user55340
13
A: White space inside XML/HTML tags

GuffaThe specification (section 3.1 Start-tags, end-tags, and empty-element tags) says that there is no white space between the '<' and the tag name, between '</' and the tag name, or inside '/>'. You can add white space after the tag name, though: <foo > </foo > <bar />

 
user55340
Thats the within <elem >value</elem> -- which is valid. As <elem>value </elem>...
 
user55340
3:24 PM
 
user55340
> Usually without DTD or XML schema definition, all whitespaces are significant whitespaces and should be preserved. However, with DTD or XML schema definitions, only the whitespaces in the content are significant
 
user55340
 
jozefg
jozefg
I'm proud of myself, Chrome now autocompletes "n" to "norvig.com" instead of "netflix.com" for me
 
Ampt
Ampt
would editing my question to add the following degrade the question:
> EDIT: I'm not really looking for guidance on this specific file, but in a more general sense. What if the classes ended up being 75 - 100 lines a piece? Would it make sense to split them then? Where is the cutoff for saying you should split a file up?
I don't want it to break down to LOC
but its a good way to measure size
perhaps I should look at cyclomatic complexity?
 
jozefg
jozefg
@Ampt Aren't python files isomorphic to modules? If so I'd just go based on what you'd want to be in a single module
 
Jimmy Hoffa
Jimmy Hoffa
3:31 PM
@jozefg Do you have a citation for this? I would have thought they wouldn't have aimed at OCaml with the syntax if they wanted anything haskell like.
 
jozefg
jozefg
I'm looking for it now..
 
user55340
Complexity is one approach... the other is the human judgement of 'does this stuff belong together?'
 
Jimmy Hoffa
Jimmy Hoffa
I understand because it's on the CLR they wanted object syntax, but they could do object syntax without the rest of the OCaml cruft like crappy pattern match syntax mixed up with guard syntax, and lack of currying
 
user55340
Breaking code into smaller modules just because "oh no, its 150 lines... must make two 75 line modules" is bad.
 
jozefg
jozefg
@JimmyHoffa This looks promising.. infoq.com/interviews/F-Sharp-Don-Syme
 
Ampt
Ampt
3:34 PM
@MichaelT Agreed, but LOC is often the lowest common denominator when relating code bases, for better or worse
Is it considered Pythonic to have multiple classes defined in the same file? http://programmers.stackexchange.com/q/209982?atw=1 #python
Tweeted by StackProgrammer on August 30, 2013 at 3:15 PM
got tweeted
w00t
 
user41796
@Ampt it wouldn't degrade, but I don't think it's necessary
 
Ampt
Ampt
I changed it to remove references to LOC
> EDIT: I'm not really looking for guidance on this specific file, but in a more general sense. What if the classes ended up being 3-5 moderately complex methods? Would it make sense to split them then? Where is the cutoff for saying you should split a file up?
 
user41796
the first answer is kind of boilerplate gimme-rep nonsense. The second answer is better, but would be stronger with an external reference or more expertise
 
Ampt
Ampt
yeah neither of them is getting my upvote yet
I've read the manifesto multiple times but it never spoke to me on classes and file size
I'm just afraid that the canon response is that its completely subjective
even though python says that there is only one correct way to do things ;)
 
user41796
@Ampt hence my comment on that 2nd answer
 
Ampt
Ampt
3:40 PM
@GlenH7 just saw that and added my own
 
jozefg
jozefg
@JimmyHoffa Ok, So Haskell.NET's failure + OCamls multiparadigm nature + .NET wanting to diversify = F#
 
user41796
@jozefg nah, just needs better marketing. :-)
 
Ampt
Ampt
averaging one view per minute. So far, so good
 
Jimmy Hoffa
Jimmy Hoffa
@jozefg Ela is the success of Haskell.NET in my book.
as much as it'll get, though I'm still annoyed to crap by the :: and : being reversed; you use cons way more than you use type hints
 
jozefg
jozefg
:P If you take success with a grain of salt
 
Ampt
Ampt
3:53 PM
I'll probably change the code on that question after lunch. they can't seem to get around the fact that i was just showing my polymorphic design more than anything
 
Jimmy Hoffa
Jimmy Hoffa
@jozefg It's successful because it works and has the basics that I would want from a haskell implementation on .NET. There's zero adoption but that's no different than haskell not on .NET
it wouldn't be haskell if it had adoption
it would be java
heh
@jozefg I'll have to watch all of that later.. good find
 
jozefg
jozefg
That's fair haha
And yeah, Don Syme and SPJ are really good speakers
 
Ampt
Ampt
4:44 PM
already have 40 hours this week... to leave or to stay and make overtime..... hmmm
 
user55340
@Ampt In situations where you've got overtime, and it isn't "encouraged" to begin with, ask your lead.
 
Ampt
Ampt
well its usually those 'take half hour lunch instead of hour' and 'end up getting to work 20 minutes early because of traffic' that just add up
and I've played it both ways before and no one has mentioned it
so I don't think it's an issue
its more of "do I want to go sit at home for the afternoon, or get paid"
but I do appreciate the suggestion
can methods have parents?
like I want to wrap a lot of methods with the same set of code
but still call the method by name rather than the wrapper and passing in the method
 
Jimmy Hoffa
Jimmy Hoffa
First person who says AOP get's stabbed
@Ampt Make a function that takes a function; does the boilerplate then executes the function passed in
or use inheritance
 
Ampt
Ampt
method inheritance?
 
Jimmy Hoffa
Jimmy Hoffa
class inheritance
a method can override the base classes implementation and then inside of it call base.SameMethod()
that's totally common
 
Ampt
Ampt
4:57 PM
oh ok, class inheritance
@JimmyHoffa is that a..... design pattern???
 
Jimmy Hoffa
Jimmy Hoffa
Nope
:)
 
Ampt
Ampt
> Leading-Edge Java
Design Principles from Design Patterns
A Conversation with Erich Gamma, Part III
by Bill Venners
as linked by @JimmyHoffa
 
Jimmy Hoffa
Jimmy Hoffa
Yeah he's one of the GoF, but he's suggesting composition over inheritance
neither of which are a pattern
 
Ampt
Ampt
who are the Gang of Four?
2
 
Jimmy Hoffa
Jimmy Hoffa
Design Patterns
Design Patterns: Elements of Reusable Object-Oriented Software is a software engineering book describing recurring solutions to common problems in software design. The book's authors are Erich Gamma, Richard Helm, Ralph Johnson and John Vlissides with a foreword by Grady Booch. The authors are often referred to as the Gang of Four, GoF, or Go4. The book is divided into two parts, with the first two chapters exploring the capabilities and pitfalls of object-oriented programming, and the remaining chapters describing 23 classic software design patterns. The book includes examples in C++ a...
 
Ampt
Ampt
5:01 PM
yes, but is there any sort of reason why they stick out more than other authors of design patterns books? or did they just do it first? Are they really so hard to name or named so often as to warrant not just a group name, but an abbreviated one at that?
 
jozefg
jozefg
@JimmyHoffa Did you see the Oreilly book on design patterns for functional programming languages?
The software engineers! They've found us!
 
Ampt
Ampt
Why is my comment about the GoF starred? Did I say somthing funny?
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt I already told you to be working through Yegge's blog sites.google.com/site/steveyegge2/ten-great-books
 
Ampt
Ampt
Gee, it must have slipped through the cracks somehow. Not sure how that happened!
 
Jimmy Hoffa
Jimmy Hoffa
;P
> Ever notice how "Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn" and "Protected abstract virtual typedef'd copy constructor function" sound identical underwater? Yup. The Elder Gods are lumbering and gibbering around Redmond right now.
It's really not hard to read Yegge, he's a very good and entertaining writer
 
Ampt
Ampt
5:09 PM
you still haven't told me why you pinned my comment :P
 
jozefg
jozefg
5:26 PM
Whenever someone says "I've got an idea for a new language" I've started to insert [bad], so "I've got a bad idea for a new language"
 
Jimmy Hoffa
Jimmy Hoffa
@jozefg I've got an idea for a new language, it will slightly corrupt some of the memory with every cycle, this way it's resistant to space radiation; that is, the space radiation has no added effect. I'll sell it to NASA and make a fortune! Viva La Ideaman!
 
jozefg
jozefg
Quick, we need a patent for an idea this cunning!
 
Jimmy Hoffa
Jimmy Hoffa
@jozefg Shit, Sun already patented this, they called it "Java" ? Weird...
 
jozefg
jozefg
Hahaha, I know, let's just throw in malloc and free! That way you can optimize for performance @JimmyHoffa
 
Ampt
Ampt
5:41 PM
@JimmyHoffa isn't there like a monad for wrapping functions and stuff
 
jozefg
jozefg
@Ampt The reader monad is the monad of a function
 
Jimmy Hoffa
Jimmy Hoffa
5:56 PM
@Ampt You don't need a monad, a functor would suffice for your purpose, but if you do so wish, maztravel.com/haskell/readerMonad.html
 
Robert Harvey
Robert Harvey
6:17 PM
@JimmyHoffa Ah, I see. That's where the deadlock issue comes in, I guess.
 
Jimmy Hoffa
Jimmy Hoffa
@RobertHarvey I'm not sure what deadlock issue you're concerned about? Deadlocks can occur in the same way as any other concurrency situation...
 
user41796
@Ampt GoF were the first as far as patterns are concerned, and the design patterns book is considered canonical for the field. That having been said, some of their explanations are kind of opaque and there are better written introductions to patterns out there. They did an amazing job of being pioneers and summarizing lots of threads and conversations in that space.
 
user41796
AFAIK, all of them were / are heavyweights in the CS realm. Having Booch write the opening for their book didn't hurt either. I believe Gamma & Booch worked together at Rational
 
Jimmy Hoffa
Jimmy Hoffa
@GlenH7 Have you ever used ClearCase/ClearQuest?
 
user41796
yes, but it was a long time ago
 
Jimmy Hoffa
Jimmy Hoffa
6:30 PM
They were both pretty damned advanced considering their contemporaries, thinking back I'm amazed at some of the facilities of ClearCase
It could function as a distributed source control before the concept was even known
ok that may be a stretch; I'm sure the concept is very old, but it still made the modern choices look like retards for all the fanbois that talk like they invented something
It's downfall was that it was mega slow and near impossible to use heh
as is so common the problem when you've got features ahead of your time
 
user55340
I'm wondering if this statement is "too profound"...
 
user41796
Rational was / is very avant garde in their thinking. RAD / RSA were amazing products when I was working with them. Honestly, I really miss the UML tooling available with RSA.
 
user55340
If a question requires 'suggested answer guidance' to keep a answers on track - its not a good question to begin with.
 
user41796
Rose (precursor to RAD / RSA / eclipse) was an amazing IDE, and it was because Rational "got it" when it came to development.
 
Jimmy Hoffa
Jimmy Hoffa
@GlenH7 They also "got it" when it came to clients money; I wonder how many businesses closed the doors after blowing half their annual budget just to get up on the Rational Suite; that was there other major downfall heh
 
user41796
6:36 PM
no doubt, their products weren't cheap. And they had strong anti-piracy measures in place.
 
user41796
@MichaelT it's a bit profound, but it's true.
 
Jimmy Hoffa
Jimmy Hoffa
@GlenH7 and that's not even accounting for consulting fees which were basically a necessity because of how complex their products were to get setup
 
user55340
I'm mildly frustrated with those questions that are closed (because they aren't good questions to begin with) where the OP is like "ohh, but if I say you have to answer it this way, it isn't too broad"
 
user41796
natural fit for an acquisition by IBM then.
 
user55340
 
user55340
6:39 PM
 
user55340
@JimmyHoffa where did you put that hit head here sign?
 
user55340
Inbound spam!
 
user55340
-2
Q: ^%#$@$Bayern Munich vs Chelsea Live%^%$%$#$@

user248034Watch Live Streaming Link perceived styles of football. Pointing to the badge on his tracksuit top, he said: “It’s white but it’s Chelsea. If you want questions about Chelsea, I’m here. If you want questions about Real Madrid and Barcelona, I’m not here. It’s not about me and Pep. It’s about Ch...

java programming-languages design-patterns learning javascript
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT I'm not posting it in here again, someone's liable to break their monitor; we're programmers after all, we have hard heads.
 
user55340
Btw, flaggers might want to hit stackexchange.com/users/3175633/user248034?tab=accounts and go to the other sites where they're a bit slower getting rid of spam.
 
user55340
6:44 PM
@Weston.h That 'set' database question... its a recommendation question now.
 
user55340
(if you assume the premise that the guy is actually asking something that he's clear about)
 
enderland
enderland
@MichaelT 6 spam flags auto deletes right?
 
user55340
@enderland Yep.
 
user20683
@MichaelT The Steaming crap is autodeleted now
 
user55340
@WorldEngineer Its still quite alive on other sites... the faster its gone, the better?
 
user20683
6:48 PM
aye
 
user55340
Anyways... that set question... its... why?
 
user20683
@MichaelT I'll leave it open for the time being
 
user20683
if it gets burned again, I'll leave it be
 
user55340
If he isn't unclear about what he's asking... he's asking for a recommendation.
 
user20683
@MichaelT It's not a recommendation question as it stands
 
user20683
6:52 PM
as far as I can tell it's are there databases that are "set" databases, the answer being "yes" but seldom by that name
 
user41796
@WorldEngineer It doesn't quite read that way at the moment. And part of the problem may be the history from the comments. It's hard for me to separate what I already knew about that question from where it is now with the edit.
 
00:00 - 19:0019:00 - 23:00

« first day (1092 days earlier)      last day (3892 days later) »