« first day (1060 days earlier)      last day (3929 days later) » 
00:00 - 19:0019:00 - 00:00

Ampt
Ampt
7:00 PM
it was like a yellow pad where you would write what you wanted to be converted into text
and it was completely seperate from the application you were using
not even converted
it would just turn what you wrote into something office could see, which it then put in as an image more or less
 
user20683
ah
 
user20683
I've not used office much in the last 10 years
 
user20683
Open Office yes
 
user20683
MS office...not so much
 
Ampt
Ampt
well it was only for pen input
 
user55340
7:02 PM
@Ampt Yep... one department doesn't like the potential success of another... and so makes it impossible for that other one to succeed.
 
Ampt
Ampt
I love MS office
 
user55340
@Ampt Read "Slack" and you might have another view...
 
user20683
I don't loathe it as much anymore. I hated the ribbon for a long while
 
user55340
 
Ampt
Ampt
I don't mean for my team or for my company or for anyone else
I use office of my own accord because I find it to offer powerful tools that help me accomplish my goals
 
user55340
7:03 PM
MS Office is a good piece of software... but its awful when it comes to "productivity". It took out the secretaries and made managers into secretaries... so now instead of having enough time to do things, they now have to write their own reports and schedule their own meetings.
 
Ampt
Ampt
everyone else can use whatever they please
 
user20683
 
user20683
some fun for you guys
 
user55340
So instead of managing, managers are secretaries... this isn't a good thing.
 
user55340
 
user55340
7:05 PM
@Ampt Oh... did you read the bit about the colored e-ink technology they had long before its time too?
 
Ampt
Ampt
not at all
they had colored e-ink?
 
user20683
They basically had all that stuff in the labs in the 70s
 
user55340
> For example, early in my tenure, our group of very clever graphics experts invented a way to display text on screen called ClearType. It worked by using the color dots of liquid crystal displays to make type much more readable on the screen. Although we built it to help sell e-books, it gave Microsoft a huge potential advantage for every device with a screen.
 
user55340
> But it also annoyed other Microsoft groups that felt threatened by our success.
 
user55340
> Engineers in the Windows group falsely claimed it made the display go haywire when certain colors were used. The head of Office products said it was fuzzy and gave him headaches. The vice president for pocket devices was blunter: he’d support ClearType and use it, but only if I transferred the program and the programmers to his control.
 
user55340
7:07 PM
> As a result, even though it received much public praise, internal promotion and patents, a decade passed before a fully operational version of ClearType finally made it into Windows.
 
Ampt
Ampt
ok, we need to stop before I gouge my eyes out
 
user20683
What are we doing again?
 
user55340
Ok, not e-ink... but a better rendering for lcd displays.
 
Ampt
Ampt
If I hear one more story of msft having implemented a million dollar idea and then sitting on it because they don't know how to manage the broad side of a barn, I'm going to lose it
who sabotages their own company.......
 
gnat
gnat
career advice, which technology to go for next, Gorilla vs Shark, all in one...
-3
Q: Should I ditch .NET C# to go with Node.js Javascript?

user96307I have been doing .NET ASP.NET webforms/MVC for a while. I have got into iOS development in the past couple of years. I'd really like to focus my career path on mobile development. A lot of my iOS apps require a backend RESTful API. I could either develop the API using ASP.NET Web API or Node.js ...

javascript .net career-development node.js
 
user55340
7:09 PM
@Ampt When the company is set up that way... they do it to themselves.
 
user20683
@gnat already dead
 
user55340
If one department gets accolades they get pay raises... and another department gets pay cuts. So you can't let the other department get a new or better technology faster than the slowest one.
 
user20683
@MichaelT who the hell came up with this management system?
 
gnat
gnat
@WorldEngineer that was fast! you did it while I was still pondering which close reason to choose :)
 
user55340
Looks like HR / upper management trying to copy other successful companies...
 
user55340
7:11 PM
Vitality curve
A vitality curve is a leadership construct whereby a workforce is graded in accordance with the individual productivity of its members. It is also known as forced ranking, forced distribution, rank and yank, and stack ranking. For example, there is an often cited "80-20 rule" - also known as the "Pareto principle" or the "Law of the Vital Few" - whereby 80% of crimes are committed by 20% of criminals, or 80% of useful research results are produced by 20% of the academics, and so forth. In some cases such "80-20" tendencies do emerge, and a Pareto distribution curve is a fuller representat...
 
user55340
> In a memo to all Microsoft employees dated April 21, 2011, chief executive Steve Ballmer announced the company would make explicit the Vitality Curve model of performance evaluation: "We are making this change so all employees see a clear, simple, and predictable link between their performance, their rating, and their compensation."
 
user55340
> The new model has 5 buckets of pre-defined size (20%, 20%, 40%, 13%, and 7%), and management simply ranks. All compensation is pre-defined based on the bucket, and employees in the bottom bucket are ineligible to move positions with the understanding they will soon be yanked.
 
user55340
(realize that by successful... and MS started this in 2006... they mean GE, Enron, Motorola, Dow Chemical, and IBM... lets look at were some of these companies are today...)
 
Ampt
Ampt
No one knows the story of Motorolas success better than I
 
user20683
@Ampt What success? They are getting eaten alive by ARM.
 
Ampt
Ampt
7:18 PM
"success" in 1997 they came to a little town where my mother grew up (about 15 minutes from where I grew up) and built a MASSIVE plant/office
> The 1,547,917-square-foot manufacturing building at 2001 N. Division St. was built in 1997 in the McHenry County Industrial submarket. The facility consists of 26% office space and 74% manufacturing, distribution and central service space.
and it was finished
and then shut down
because they couldn't afford it
and hasn't been used since
drive by quite often, and its just astounding how huge it is... and that no one ever even used it
scratch that, they used it for a short while
still hit the locals pretty hard to lose 2.5k jobs
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT This is a trend that I have found baffling, but then, maybe when managers had secretaries, they did even less than they do now? That's an even more baffling thought...
@Ampt Like I said earlier, they are very vertically segmented, and there are lots of segments..
 
user55340
@JimmyHoffa Its not a "more / less" its a "managers are now secretaries" - so (making up numbers) instead of having a secretary at 20k/year doing things a manager is spending more time at 100k/year to do things that the secretaries used to do... because they have 'slack' in their schedule (title of the book).
 
user55340
But its that slack that is used to handle emergencies, work with employees and the like. And that has largely gone away. So any emergency requires extraordinary measures.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT Ah, so you're saying they used to just have that time as extra for emergency events? (Did they do something in emergencies back then? Last I heard they still just make phone calls during emergencies)
 
user55340
7:33 PM
(hmm... I gave away my copy of Slack, and its not in Safari Online...)
 
user55340
It was an attempt to get my previous employer to wake up. I gave away copies of Slack, Peopleware, and Death march to various managers and directors.
2
 
user20683
did it waken them or piss them off?
 
user55340
Unfortunately, neither.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT That's awesome. I've tried similar approaches before, sadly these types of people who have those problems are the same type who would never read or study anything anyway because they think they're doing fine, and/or frankly just don't care about the office, that's just work; they live for other stuff
 
user55340
They offered to reimburse me, my comment was "If there are any changes because of this, then I am reimbursed. If there aren't, it doesn't make too much of a difference in the long run."
 
user55340
7:37 PM
(the long run being "I'll get a better job that pays more that I enjoy")
 
user55340
Throughout Deathmarch there is an underlying theme of "Morale is important, and quitting is always an option."
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT You know they never read them, right?
 
user55340
I know the director who was demoted to manager shortly afterwards was reading them.
 
user55340
I think my manager was reading Death March (I think I saw the bookmark move once).
 
user55340
I also think that the HR (who gave her notice the week after I left) had ordered at least Peopleware...
 
user55340
7:41 PM
So... I'm not sure if I was reimbursed or not.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT Could either of those people actually do anything to solve any of the problems the book presented? If not, and they did read them, then you're just literally torturing them
 
user55340
The director at the time and HR should have been able to make changes. If they couldn't (and this is probably the case) there are bigger problems further up in the chain.
 
Jimmy Hoffa
Jimmy Hoffa
"Everything we're doing wrong, right here, in black and white, with clear and concise solutions, all we have to do is X, Y, and Z, but we won't. It will never happen, and I can't make it. I can't unsee, augh...."
@MichaelT have you solved that self-referential test?
 
user55340
@JimmyHoffa My father has solved it... its very not-trivial.
 
user20683
It's solvable, that much I can figure out
 
Jimmy Hoffa
Jimmy Hoffa
7:47 PM
@MichaelT Yeah, I realized this just now as I noticed the q/a each are individually evaluated so the correct answer changes
20 is just the seed for the algorithm
it needs that one fixed point
otherwise it would have far too numerous possible solutions
 
user55340
 
Jimmy Hoffa
Jimmy Hoffa
 
user55340
My personal favorite question on the Drunk Men test is #19.
 
user55340
19. The answer to this question is:
A O A
B O B
C O C
D O D
E O E
 
Jimmy Hoffa
Jimmy Hoffa
The "With a unique solution" I was thinking it meant interesting, but now I realize what it meant.
 
Jimmy Hoffa
Jimmy Hoffa
8:05 PM
@MichaelT This is exactly the sort of thing calculus exists for isn't it? This would be a beautifully prolog problem
 
user55340
@JimmyHoffa Don't need calc for it. Prolog would probably tackle it.
 
Jimmy Hoffa
Jimmy Hoffa
This could give me exactly the problem to genuinely write some prolog with in fact, think I'll do that
 
user55340
@JimmyHoffa Given that there are other tags of "where the problem came from"... maybe you need to make a tag on SO for any questions related to this of drunk-men-work-here?
 
user55340
Just think of all the fun you could get in on MSO debating if that tag should exist.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT Why do I have this sense that calculus is for deriving the values from these types of scenarios? Isn't there like some standard common calculus people are taught that is for exactly this kind of thing?
 
user20683
8:12 PM
@JimmyHoffa what sort of thing?
 
Jimmy Hoffa
Jimmy Hoffa
@WorldEngineer That self referential test, it seems like one of those word problems that I think are a part of calculus where you have multiple variants and based on their relationships have to come up with a model that fits all the relationships to result in giving you usable values for those variants?
 
user20683
That might be nonlinear optimization, I can't recall
 
user20683
brain is a bit fried at the moment
 
Jimmy Hoffa
Jimmy Hoffa
I've never taken calculus so iduno, I just have these vague senses about what subjects some things fit into, no real idea though
 
user20683
It sounds almost like linear programming
 
Jimmy Hoffa
Jimmy Hoffa
8:17 PM
from a programming sense obviously any constraint based things will solve the problem I recognize that; prolog, that linear programming thing you showed the other day, I know there is a calculi of constraints I've seen reference to, but I thought some of this stuff was much more basic calculus
more common maths stuff that isn't so focussed on programming has well-known solutions to this type of problem (I think?)
 
user20683
integrals are kind of like that, if you do several of them, they enclose an area.
 
user20683
Analytic Solution(s) is the "proper" term
 
user55340
@WorldEngineer I think it is most similar to linear programming... though some of those are difficult to express.
 
user20683
Is a possible option as well
 
user20683
Nonlinear programming
In mathematics, nonlinear programming (NLP) is the process of solving an optimization problem defined by a system of equalities and inequalities, collectively termed constraints, over a set of unknown real variables, along with an objective function to be maximized or minimized, where some of the constraints or the objective function are nonlinear. It is the sub-field of Mathematical optimization that deals with problems that are not linear. Applicability A typical nonconvex problem is that of optimising transportation costs by selection from a set of transportion methods, one or more of ...
 
user20683
8:21 PM
there are days when I wish I had sunk a extra 30 hours or so into a proper math degree
 
user55340
I think it is linear... with 20 equations and 20 variables.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT When is a problem lineaer vs nonlinear? (I never could understand linear algebra, too much "This just works, because magic so shutup")
 
user20683
Linear is a function with a degree of 1 or less
 
user20683
x + 2
 
user20683
non-linear is higher degrees
 
Jimmy Hoffa
Jimmy Hoffa
8:25 PM
"degree" if nothing else I wish I had a math->programming lexicon
 
user20683
sorry
 
user20683
I actually have that wrong
 
user20683
linear = proportional
 
user20683
so linear would be where the output is directly related to the input
 
user20683
nonlinear is everything else
 
user20683
8:26 PM
like Logs and Sines
 
Jimmy Hoffa
Jimmy Hoffa
ohh
that makes sense
 
user20683
so x^2+x-5 is linear
 
Jimmy Hoffa
Jimmy Hoffa
like in big-O, N is linear
 
user20683
yeah
 
Jimmy Hoffa
Jimmy Hoffa
makes me think about polynomial vs non-polynomial
 
user20683
8:26 PM
exactly
 
user20683
I take that back again, x^2 is not linear
 
user20683
ugh
 
user20683
this is why I shouldn't explain math on 1.5 hours of sleep
 
Jimmy Hoffa
Jimmy Hoffa
In that case, @MichaelT I disagree, I think it's non-linear because the relationship graph is wildly variable wrought with deadends that you couldn't diminish without testing
I think it's NP-Hard
@WorldEngineer That's ok, I didn't even see where you said that, the relationship to Big-O made it make perfect sense
 
user55340
Try writing up the equations for them... sometimes they simplify. (not that this has anything to do with linear or not)
 
user20683
8:31 PM
so linear is 3x+ 4y + 7z = 15 say
 
user20683
while non-linear might be x^2-log(9y) + sin(z) = 5
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT sure, I could see that, but there are many that are more than one point away, that indirection involves undecidability
That test could be linear, if the questions ensured relational dependencies only on immediate family and never extended family, but that particular implementation of that test is not as such
 
user20683
@JimmyHoffa there's a guy at MIT who is amazing at explaining math
 
Jimmy Hoffa
Jimmy Hoffa
Things like "How many questions have consonants" combined with a couple others that are globally quantified are where you run into needing to test all paths
 
user20683
Gilbert Strang
 
Jimmy Hoffa
Jimmy Hoffa
8:34 PM
@WorldEngineer I first read this as "there's a guy at MIT who is amazing at math" and had a good larf
 
user20683
well there is
 
user20683
:P
 
user20683
Gilbert Strang
William Gilbert Strang (born November 27, 1934 in Chicago), usually known as simply Gilbert Strang or Gil Strang, is an American mathematician, with contributions to finite element theory, the calculus of variations, wavelet analysis and linear algebra. He has made many contributions to mathematics education, including publishing seven mathematics textbooks and one monograph. Strang is the MathWorks Professor of Mathematics at the Massachusetts Institute of Technology. He teaches Introduction to Linear Algebra and Computational Science and Engineering and his lectures are freely availab...
 
user20683
His book on Linear Algebra is available for free IIRC
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT Am I convincing yet that it's non-linear, or am I as I figure, missing something you recognize about it?
 
user20683
8:36 PM
oh and the Calculus you were thinking of is called the "Calculus of Variations"
 
user55340
Not sure... its been a few years since I attacked it.
 
user55340
To me, x + y + z = 5 or is(A,q1) + is(A,q2) + ... is(A,q20) = q2 [A: 6, B: 7, C: 8, D: 9, E: 10] isn't too different.
 
Jimmy Hoffa
Jimmy Hoffa
@WorldEngineer I guess it's a bit more than the simple basal calculus people learn like I thought..
 
psr
psr
@JimmyHoffa Prolog would be fine - it appears (if I didn't miss anything) that first order predicate logic handles it fine.
 
Jimmy Hoffa
Jimmy Hoffa
@psr Yeah prolog should do it perfectly I agree. That's exactly the type of thing prolog is for.
maybe it is linear. Iduno.
 
psr
psr
8:47 PM
I used to know someone who was very practiced at first order predicate logic and would just write things like that test down on paper and whip through the solution.
 
user55340
Btw, there is some other rather neat stuff on the site - drunkmenworkhere.org/games (circling the square is rather neat)
 
Jimmy Hoffa
Jimmy Hoffa
@psr That's really cool. What makes a predicate "first order" vs higher order?
The number of globally quantified predicates are what really throw me on that
> 9. The next question with the same answer as this one is question:
> 3. The number of questions with the answer E is:
I guess it's not globally quantified when you're given 4 choices, so they rely on those 4 questions which may have a mutually recursive dependency though is what throws me
 
user55340
#9 is rather neat because it constraints other ones. Either 10 is A, or 11 is B, or 12 is C, or 13 is D or 15 is E.
 
psr
psr
@JimmyHoffa I'm thinking you have to convert those into first order statements based on some horrible and/or conditions.
 
user55340
And there are a bunch of nots in there too.
 
psr
psr
8:52 PM
Like MichaelT just did.
 
Jimmy Hoffa
Jimmy Hoffa
@MichaelT Which due to the dependence on 15, it depends on a globally quantified one: 
12. The number of questions whose answer is a consonant is:
Aan even number
Ban odd number
Ca perfect square
Da prime
Edivisible by 5
 
user55340
Da Prime!
 
Jimmy Hoffa
Jimmy Hoffa
heh yeah
And while being globally quantified, it has other dependants because prime and odd overlap so if it's one of those two, which depends on and regulates the answer to another question
 
user55340
#12 is even more fun... I think its either A or B, because the other ones create multiples... though I'm not sure.
 
Jimmy Hoffa
Jimmy Hoffa
A and E also overlap
actually E overlaps with A and B
and D
 
user55340
8:55 PM
Well, D would have to be extrordiary to overlap with B.
 
Jimmy Hoffa
Jimmy Hoffa
no not D nevermind
2 isn't extraordinary
it's a totally common number :)
 
user55340
Only having 2 consonant answers would break many other things.
 
Jimmy Hoffa
Jimmy Hoffa
that's exactly my point
I'm convinced it's non-linear
 
user55340
Logical inconsistencies isn't about linearness.
 
Jimmy Hoffa
Jimmy Hoffa
you have no idea how many things it would break or fix without trying it
@MichaelT my point is to evaluate the validity of a solution against a constraint requires multiple constraint evaluations
 
user55340
8:57 PM
So? Even propositional logic handles that.
 
user55340
(btw, #19 is not A because of #13)
 
Jimmy Hoffa
Jimmy Hoffa
...I don't know what that is...
 
psr
psr
21. How many nerds did this web page successfully snipe?
 
user55340
Prop logic is simple "and", "or", "therefore" type truth tables... logic gates.
 
user55340
(the neat part of the test is that when something is wrong it highlights it)
 
user55340
8:59 PM
Though I wish you could mark something as impossible to help narrow it down online.
 
Jimmy Hoffa
Jimmy Hoffa
but I'm still convinced, if evaluating constraint A requires evaluating B, C, and D, and evaluationg B requires evaluating E, F, and G, and evaluating E requires evaluating A, D, and F, then the expanse of any individual constraint is no longer linear; each additional question increases the search space for any of the global quantifies by more than 1
especially due to the recursion at the end where everything might evaluate fine with A being bla until a dependency of it which depends on it doesn't work when A is bla
 
user55340
 
psr
psr
I think linear means (in this case) that you have ands, ors, and nots. I seem to recall that every, some, and none are second order, or maybe you get something else at that point. Nothing to do with big O.
 
user55340
Thats the classic propositional logic type thing.
 
Jimmy Hoffa
Jimmy Hoffa
@psr Right, that's my point, and there are "every" type constraints in there
12 is a some constraint
 
user20683
9:02 PM
there's prop logic and then predicate logic
 
user20683
which is "2nd" order IIRC
 
psr
psr
@JimmyHoffa True, but I thought you could replace them by exhausting all the cases.
Note that I've been too lazy to look any tiny part of all of this up or anything.
 
Jimmy Hoffa
Jimmy Hoffa
@psr exhausting all the cases is what every/some/none are and what makes them beyond the scope of linear/first order/etc
@psr I'm guessing as much as you. More probably, I am sincerely undereducated to speak on this topic
 
psr
psr
@JimmyHoffa I know - but with only 20 questions you can convert "any" to 1 or 2 or 3 or 4 ... or 20. "Some" is combinatorially ugly, as is the number 4 (1 and 2 and 3 and 4 and not 5 and not 6...or not 1 and 2 and 3 and 4 and 5 and not 6...)
 
Jimmy Hoffa
Jimmy Hoffa
@psr I think you pegged it with question 21.
 
psr
psr
9:19 PM
Actually, I'm wondering what the test author thinks the answer to 20 is.
 
user55340
I'm fairly sure its 'E'
 
Jimmy Hoffa
Jimmy Hoffa
@psr Yeah, I had the same idea; because there's still a relationship with barometric pressure and all of those things except for longitude, but he means to say there is no relationship which is a bit false
 
psr
psr
Or that it vaguely sort of ties into a lot of stuff except sometimes it's just not related.
 
Jimmy Hoffa
Jimmy Hoffa
@psr That would make sense except that he threw longitude in there. Barometric pressure does not have a correlation with longitude, all the other things it does
 
Jimmy Hoffa
Jimmy Hoffa
10:04 PM
-1
Q: Where can I find some good Ruby on Rails themes for a small portfolio type site?

Louis93I am open to themes made for Jekyll as well. Well designed themes seem to be particularly hard to find, I was wondering if you guys had some better links...

web-development ruby-on-rails
 
psr
psr
@JimmyHoffa Right, which could just mean he thinks some of the questions on an IQ test don't correlate with intelligence.
 
Jimmy Hoffa
Jimmy Hoffa
ah
Neat. That actually makes scala not seem so dirty. Or rather, it teaches you to write Haskell in Scala, which is kinda wrong for it's own myriad of reasons.
 
psr
psr
@JimmyHoffa Spent time over the weekend going over fixed point combinator - mostly the answer to your bounty question. Sort of like first seeing recursion "oh the loop is hidden in the stack", except now it's "the loop has been turned inside out and it's spine has been stapled to the stack".
 
Jimmy Hoffa
Jimmy Hoffa
@psr Heh yeah, you see why I said I start getting a headache and my vision goes blurry anytime I try to use indirection that works anywhere near fixed point recursion
 
psr
psr
@JimmyHoffa I need a Haskell phrasebook for humans.
@JimmyHoffa My goal was to get some intuition about it, and I think I had some degree of success. But the details are more intricate than looping or recursion, even for a "simple" combinator.
 
Jimmy Hoffa
Jimmy Hoffa
10:18 PM
Just the simple act of trying to mentally contort myself to derive the Y combinator on my own in a few different languages has been a learning experience on it's own, I tried quite a few times for a good while without ever succeeding. Learned to smell it coming though and how to recognize more quickly when I'm approaching an infinite type definition problem
Yeah, some intuition about it is all I've managed, I'm sure if I read up on it more thoroughly than just trying to force myself to grok it in fullness I might have better luck; but then I wouldn't enjoy it as much heh
There is definitely clear and present reason for a company trying to have cred as a group of intelligents to name itself Y-Combinator and reason for it to be on the MIT scheme logo, it's no joke.
 
Jimmy Hoffa
Jimmy Hoffa
10:42 PM
@gnat that's a pretty good read translating haskell to scala, but not showing TypeClasses is kind of like comparing a functional language to LISP and not showing macros just because the other language doesn't have it. This clarifies a lot here to show Scala is really just OCaml with a non-ML syntax. The thing that annoys me about it is I just can't explain why it didn't use the ML syntax, but I guess it doesn't matter, it all seems exactly the same (traits being the one extra thing Scala has)
Traits are roughly similar to typeclasses though and can probably kind of stand in
@WorldEngineer helm-engine.org
 
Jimmy Hoffa
Jimmy Hoffa
11:04 PM
Very interesting; some fellow says his high school students doing scratch and other imperative things were totally disinterested, but after trying some things for years, he did haskell for 2 years and found them far more interested because it wasn't thinking step-by-step like a machine which is more foreign to the kids (or so is his hypothesis anywho) sawafaso.blogspot.com/2013/07/…
0
Q: What are the prerequisites for learning PHP?

LabiA professor told me that if I learn another language first such as Java it would be better because I will grasp the OOP princples better than if I start with PHP and after that I can move to PHP. Do you agree with him? Why? Why not?

php
A super-standard sized bottle of tequila and a hammer that doesn't work?
 
00:00 - 19:0019:00 - 00:00

« first day (1060 days earlier)      last day (3929 days later) »