« first day (1517 days earlier)      last day (3472 days later) » 
00:00 - 19:0019:00 - 23:00

7:00 PM
exit
 
user55340
 
your cat looks tired of this "computer" shit, and requires demands attention immediately.
 
user55340
That's possibly part of it. Though right now, his purr motor is going full power.
 
It's a good thing that his purr-train has that 10 Year / 100,000 scratches warranty then
 
user55340
he is really good about not scratching anyone intentionally. He's a happy cat.
 
user55340
7:06 PM
The only improper scratching he does in the house is the corner of the mattress jumping up on the bed. He's really a very good cat.
 
user55340
That said, he's not helping me write the 'thank you' letter for the interview.
 
Hot tip: Cat scratching furniture? Easy training solution works super well: Double sided tape wherever it normally scratches.
Used this for multiple cats, they quickly learn it's not fun to scratch there because it sticks to their paws and then they stop trying altogether
 
user55340
he's scratching the corner of a cheap mattress. He doesn't bother with the more expensive couch.
 
user55340
If I was more concerned about it, I'd build a custom scratching post for the corner of the bed... he likes scratching posts too.
 
@MichaelT was just a general all-around tip for anyone, works really well and is super easy. Just take the tape off after a few weeks and they don't go back to it.
...now if only I knew how to get our shitty angry cat to stop peeing on my kids toys out of jealousy.... yes, we do love the kid more than you, that's in no small part due to him not peeing on the carpet...
 
7:19 PM
Shameless self bump
0
Q: Data Access Layer with Asynchronous services like MQ

maple_shaftI am tasked with creating a development, design and architecture guide for a large multi-year project. I have to dictate best design practices for a number of architectural perspectives. See the below diagram to demonstrate how I see this looking. So based on other architectural mandate, MQ ...

I really have no shame
@MichaelT You look exactly like I imagined you
Like a skinny version of Kevin Smith
 
With a beard
 
you beat me to it.
 
@MichaelT WOW
That is like hitler without the stache...
 
only less... evil
 
7:22 PM
or Worf without the sash
 
user55340
@maple_shaft sash? or stash?
 
user55340
As an aside, the '#yoga hosers' appears to be:
 
user55340
> According to IMDB, "Yoga Hosers" is a film about "Two teenage yoga enthusiasts team up with a legendary man-hunter to battle with an ancient evil presence that is threatening their major party plans."
 
user55340
o_O
 
7:27 PM
@maple_shaft imagine a system where there is no synchronous calls, and nothing ever returns a value. How do you get responses? (I don't know that I entirely understand your Q? though it sounds like you're trying to do the traditional submit-asynchronously, then synchronously poll? either way, think on what I mentioned as an exercise. There's a clear approach forward in such a system, which is occasionally a good solution to use.)
 
@MichaelT You must not get out much.
They've pretty much played the genre to death, but I'll be interested to see Kevin Smiths take on the yoga enthusiast
I could see posts being async but gets being sync
 
Tucker and Dale vs. Evil is an awesome movie. Just thought I'd throw that out there...
@Ampt you're thinking in terms of REST conceptually, that's a subspace of his design problem and only a minor implementation detail
 
I'll admit, reading is hard.
 
I have to say, @maple_shaft your question doesn't seem super clear.. there's lots of detail, but the specifics of what you are having difficulty with I'm not entirely seeing
 
My guess? Terrible performance, battery life, experience in exchange for modularity...
Not that the idea is without merit, but forcing every part in the phone to have it's own protective casing and communication standard is going to be bulky at best
 
user55340
7:39 PM
Ok, back to desktop. My cat has decided that since the sun came out, the north facing window isn't warm enough and the south and west facing ones in the room with the desktop may be more optimal for cat warming.
 
user55340
@Ampt And back in the 3.5" days, we thought that 6" phones were ugly.
 
user55340
 
Highly modularity is great for some things, but I don't see it useful for a phone.
 
user55340
@ThomasOwens Have you seen the ideas behind the modular portable computing device?
 
@MichaelT :\
 
user55340
7:41 PM
One such attempt: phonebloks.com/en
 
user55340
 
@Ampt I love the idea but yeah it's going to be a magic trick to pull off... that said, I could see some things that could help it work not dissimilarly to how stuff works now- ensure the endoskeleton has enough channels in each slot to allow combined chips, so you can buy an audio/wifi/3g/SSD controller chip in a single chip
 
@MichaelT this is my favorite part of the iphone6+
I remember apple fanatics just ripping on android for large phones
 
user55340
Incidentally, that phone blocks thing was a year ago today
 
it's that multiple-features-per-chip that makes the mobile stuff work like it do.. the other issue though is then the OS makers need to handle multiple possibly underlying hardwares which is really probably the largest killer issue- though windows and mac have solved this on desktops, no reason now that modern mobile tech has enough ram (possibly significantly more than a normal mobile device in one of these componentized devices as componentized devices tend to allow higher end components) they may be able to allow more device drivers in memory
 
7:44 PM
I'm still waiting for the phoneputer
like the ubuntu phone
the ubuntu edge
 
@MichaelT sometimes these things aren't coincidences though they seem like it; somebody knows that and posted this phoneblocks thing somewhere or talked about it with someone, some people got curious started digging around and links about this stuff started showing up on various reddits or HN and before you know it techies around the web are talking about the concept today... basically you just completed the loop.
 
user55340
@JimmyHoffa I poked at it a bit ago...
 
...though perhaps it is a coincidence in this case shrug
@MichaelT I just mean the fact that we're looking at this one year to the day
 
user55340
True. Just noticed that though when viewing the video.
 
Hmmm. so interviewed this morning, not sure how I feel about the job. The benefits for it (working from home, interesting subject material) seem to be pretty awesome, the main drawback is the work might be boring or repetitive, but the manager is pretty great and seemed open to someone taking those responsibilities and running with them
 
7:50 PM
while(repetitive){automate();}
 
@Ampt see I find that problem of repetitive/boring tasks basically says "do you want to make significant improvements? yes? great!"
 
yeah if they're willing to let you spend the time to remove the repetition with tool building then I'm all for it.
 
We had a really interesting post-interview conversation, heh
I don't do well with "you must have a shitty job indefinitely" situations - I am very consistent and persistent in finding better ways to do things. I didn't want this to be unclear to the hiring manager :)
 
@MichaelT Doesn't even look like Kevin Smith.
 
user55340
@RobertHarvey Looks like a 'goofy spaz'.
 
8:37 PM
tick-tock goes the clock. hurry-up says the sock.
 
@Ampt just quit
 
@enderland Solid Workplace advice, thanks!
 
@Ampt thx
 
"How can I tell my boss I want to quit for the rest of the day, but I'll be back tomorrow morning?"
 
I just left after I got back from my interview today, so I'm going to go with "just leave"
 
user55340
8:49 PM
Ohh! Gold badge! copy editor
 
user55340
I join the elite group that includes Peter Mortensen on P.SE.
 
@MichaelT grats!
 
@MichaelT I'll have you know I just drew up a Huffman encoding tree to make sure that you had a correct encoding.
 
user55340
@Ampt Heh. I wanted to make sure I had the right info there.
 
user55340
Thank you for checking though.
 
8:53 PM
well I still think you're cheating :P I'm not sure he's storing it at the bit level
but who knows
 
user55340
Its stored somehow
 
user55340
And btw, I copied it from:
 
user55340
 
We always just made them by hand
make a tree, group the smallest two
that now has the value of both it's pieces
combine the next smallest two, repeat, etc.
 
user55340
I was able to eyeball the frequency table.
 
user55340
8:55 PM
The thing is, if they're ints... its even more savings.
 
No, I agree, it's a lot of savings
 
9:12 PM
@MichaelT ...like Kevin Smith?
@Ampt the things I should have learned in uni but instead learned by weird ass industrial use and didn't even know it had such a name until years later...
 
psr
9:28 PM
@MichaelT SE really needs a gold-badge montage app, that shows animated before-to-after scenes of the activity that earned the badge. Obviously much more impressive for badges that require doing some meritorious activity many, many times.
 
user55340
0
Q: Developing a "time scheduler" app

user3834459I need to develop an application to generate, given a set of preferences and constraints, a time schedule for employees. These tables can really get complex and they might need to take into account multiple variables (e.g. an employee needs to do a number of hours in a division and a number of h...

 
user55340
Could I get a close vote dup to:
 
user55340
2
Q: Algorithm for mapping users to a schedule based on time availability

webnoobI have a project where I must map employees to a schedule. The basic concept is: Company Opens between 08:30 and 17:00. Employee 1 can work from 08:30 to 12:00 then from 13:00 to 1700; Employee 2 can work from 12:00 to 13:00. Employee 3 can work from 17:30 to 00:00 I would then want the emplo...

 
psr
@MichaelT yes
 
user55340
@psr Thank you.
 
9:38 PM
@psr I was thinking this morning on my way into work, and you may actually want an applicative for your code generation; not a monad. Applicatives are great for partial constructions which would allow you to construct bits and pieces of your AST step by step with logic between the steps of function application
A simple way of seeing that is to realize the ap function indicative of an applicative has a sig of F (a -> b) -> F a -> F b, so given that what you can actually do is read it as F (a -> (b -> (c -> d))) -> F a -> F (b -> (c -> d)), just filling in terms one at a time doing partial construction/application with logic between each of your steps (logic gets to live in a generator of your F a in the above example)
 
psr
@JimmyHoffa I've looked at that earlier link some (in between reading about node.js, Angular, and the application that implements the low level link between node and a MUMPS database). I've been thinking more about what I can implement in MUMPS and what the AST should be. Also, I need to look up how to implement nested lexical closures. It can't be too easy since it took a while historically.
 
and you still get composition because the result is F (b -> (c -> d)) which can be handed to ap again along with a F b to compose your next step
 
psr
I also need to figure out what control structures to have. Since I'm hoping to have a javascript builder rather than a parser, for loops, while loops, and if/then might be awkward. I'm considering S expressions, but then that might be to weird for those I want to use it. But "too weird" might be unavoidable.
 
@MichaelT At Employer^^ I was once asked to come up with a scheduling algorithm for an educational institution. I even gave a time estimate. That was before I understood that scheduling algorithms can be a graduate-level problem.
 
user55340
@RobertHarvey I had a 'best first search' in AI class. It was a 'simple' job shop problem... and pointed out that many others are much more complex when additional constraints are added.
 
user55340
9:50 PM
My father would joke that scheduling programs are just conflict finders and don't actually give schedules.
 
Yeah, that pretty much sums it up.
Imagine trying to automate the scheduling of classes at a semi-large institution (2000+ students, four to six classes per day each).
 
@psr yeah the AST model and AST->MUMPS generator on it are the most important things to start with; putting an applicative or monadic or <whatever-makes-sense> front-end interface for creating the AST can be done later and will be much easier to come up with after you've got the middle representation and back-end generator ironed out. Will be less variables to juggle in your head, sounds like you've got more than enough unknowns to juggle right now...
 
psr
@JimmyHoffa Sounds plausible. So to the programmer using this, they get to write a -> (b -> (c -> d)), which is pretty normal for them, and can pass it all to the applicative, which can build an AST by breaking it down into F a -> F (b -> (c -> d)), and knowing what to do with F a and how to chain that to the remainder of the expression. I would need to look up dumbed down examples.
 
@psr yeah. A javascript interface might look like..
 
@RobertHarvey how did they do it before computers?
 
psr
9:54 PM
@JimmyHoffa I think I've decided except for control structures, and that depends a lot on how the front end interface deals with them. Basically, no point in giving people a nice familiar for-loop if the interface makes it as weird as, say, an S expression anyway.
 
@MattGiltaji Hollerith cards, folding tables, hundreds of students and dozens of teachers in a gym.
Nowadays they put it all online, and let the students sort it out ("crowdsourcing"). It actually works very well that way.
So long as the server doesn't crash.
 
user55340
Back just before I started college, it was a card system... a given section had N cards printed out. Students would get the physical card to register for the class.
 
psr
@JimmyHoffa So far it's just like what mine looks like.
 
user55340
At some point, you took the cards you had to the registrar, gave them to person working and said "I'm taking these classes, my student ID is..."
 
if.with(
  equalityComparison.with(someVarA, someVarB),
  statementBlock.with(
    print(someHappyShit),
    return(whatever)
  )
)

giving you:
{
  nodeType: 'if',
  leaves: [
    {nodeType: 'conditionalEquals', leaves: ['valOfSomeVarA', 'valOfSomeVarB']},
    {nodeType: 'block', leaves: [ /* print node, and return node */ ]}
  ]
}
 
9:59 PM
@MichaelT very interesting
 
psr
@JimmyHoffa That actually seems do-able.
 
where with is your ap, and if is a function that takes a conditional param and a then param and with applies the params one at a time (in JS a function can have it's .length checked to see how many params are left, so your with could do partial application returning a new function until it sees only 1 param is left which complets the curry because you don't have auto currying that automatically turns a -> b -> c into b -> c)
 
psr
@JimmyHoffa O.K. Having the statement block just keep adding arguments to with is clever. As a style point I'd like to avoid having people wonder why they keep having to add "with" all the time, so I'd probably call it "LongLiveTheEmperor". But otherwise it looks pretty usable and not at all bad to implement.
 
10:14 PM
yeah the with let's you say "well, I don't *know* my body of my if here, but I know the conditional, so I can `if.with(condition)` and return the result from this function and let some other piece of the code `.with(theBody)` which will cause the with to apply the function. Also if you did an applicative you would want to define your `F` -> partial construction makes since using an `Either` applicative when you want validation, any point the `F` goes to `Left "that years invalid, dolt!"` and the construction just completes with that. In your case I don't know what would make sense there, ma
> For example, State s is an abbreviation for StateT s Identity.
The identity functor/monad/whatever does nothing, just arbitrarily carries values through the compositions the interface allows. It could be thought of as a list functor/monad with only a single element ever.
as for the naming of "with" - that stuff's always tricky. It's common for people to implement kleisli in other languages with the name "then" because it makes a bit of sense, what a good sensical term for the behaviour of ap is, I don't know. I'd almost say Apply but that's already taken in JS...
it's always a trick because in Haskell you can just say kleisli or ap, in other languages you'd be strung up for naming your functions such things...
 
user55340
10:41 PM
Whee! Repcap for the day. 34 days so far...
 
00:00 - 19:0019:00 - 23:00

« first day (1517 days earlier)      last day (3472 days later) »