« first day (984 days earlier)      last day (4005 days later) » 

psr
12:07 AM
@WorldEngineer - Well, of course. They parsed using state machines.
 
user20683
@psr I was read into that one.
 
2:03 AM
@WorldEngineer so when's your first interview?
 
user20683
@JimmyHoffa unscheduled as yet, got a few more things to do first. Been quite sick all last week, set my time table back
 
user20683
ran myself ragged during finals week then we got hit by a cold spell
 
user20683
did me in nicely
 
Oof. So do you have a double bachelors or just almost?
 
user20683
@JimmyHoffa it'll be totally official on Friday
 
user20683
2:10 AM
I went to the recognition ceremony on Friday, hopped up on Dayquil but I made it through
 
user20683
I paid for it over the weekend
 
user20683
developed a horrible post nasal drip and cough
 
user20683
though at least my fever was gone by Saturday
 
Aye, am I remembering wrong in thinking you said you would have enough credits for both bachelors, or just really close to it?
 
user20683
@JimmyHoffa I had enough credits
 
user20683
2:13 AM
that post of my grades and being done was correct
 
Cool. Double bachelors is pretty cool. Plus gets you in line for your masters really well if you want one (though don't kid yourself, you're never going back, when you've been in the industry for ~1 year you'll realize this)
 
user20683
in my case it's much more likely
 
user20683
masters are more common than bachelors in my family
 
user20683
but we shall see
 
Call me crazy, but you seem too pragmatic for all that, you're going to find the job of software engineer quite comfortable, a good SE job is all about pragmatism, you'll enjoy it.
(unless you get a bad SE job, then you'll be applying for your masters in ~9 months)
 
user20683
2:28 AM
@JimmyHoffa I won't call you crazy but I also won't be applying for a masters for a while either
 
I wonder if I could compile my machinad.js/parsad.js in JScript.net to get a dll I could reference from C#, parser combinators in C# would be much trickier...though perhaps I should write them in C# using the DLR just to get a feel for how the DLR works
 
user20683
but yeah, I characterize myself as a Romantic Pragmatist
 
user20683
@JimmyHoffa I do have a couple of holes in my education: Formal Database Theory and Networking
 
user20683
but I figure I can learn both of those very quickly
 
@WorldEngineer I can't really speak with experience to this; but I find it extremely hard to imagine someone who enjoys engineering and is successful in industry even for a short period being willing to go back to school
 
user20683
2:34 AM
@JimmyHoffa regardless, it's something that I'll consider down the road.
 
that's where I learned everything I know about networking
 
user20683
@JimmyHoffa how easily does it transfer to Windows?
 
user20683
not that I'd need to that given that I don't have a working windows box at the moment but still
 
user20683
I'm definitely getting the Algorithms Design Manual
 
user20683
might pick up the gang of four as well
 
2:43 AM
@WorldEngineer you know the algorithms design manual is a free PDF online right? The UNP is very low level, it's all C coding and details how packets are made up and etc etc
 
user20683
@JimmyHoffa ah, huh
 
user20683
as far as coding, I know a decent amount of C and I could really, really use more practice on actual problems rather than "toy" ones.
 
crap I hate that google has changed so copying one of their links has frigging garbage google redirector rather than the link you are trying to copy
just google "the algorithm design manual pdf" comes right up
the UNP goes over how networking works at the low level and how to code it, it's fundamental and how it works under windows just the same as *nix. Though the examples work directly in nix, in windows you have to use winsocks to get the exact same API's (almost identical anyway, slight variations) back in the day. Winsocks is no more now though, I'm sure theres a totally different API in windows for the low level sockets coding these days, though cygwin and others will likely have the old APIs
 
user20683
@JimmyHoffa once I understand the theory, it'll be easy enough to adapt anyway
 
user20683
that's one thing that I've noticed is that I find myself pulling solutions out of thin air much, much more easily in the last year or so.
 
2:50 AM
Yeah, the fundamentals are really the same; the key is the protocols of TCP/UDP are the exact same on windows or nix, and the UNP basically goes over how to put together the packets in those protocols from scratch, and they have the same structure in either so ...
 
user20683
@JimmyHoffa it'll be nerd cred if nothing else
 
Yeah, it really doesn't come in particularly handy, but to understand what you're looking at when debugging stuff and what to expect when trying to put together something network based, you'll know you need to make the socket listen somehow and fork somehow et al and won't put together a solution missing those things and wonder why it doesn't work
 
user20683
@JimmyHoffa also assuming I go into web dev work, it's good to know what's going on behind the curtain.
 
@WorldEngineer Classic web dev interview question to verify fundamentals: Name all the HTTP verbs you can
 
psr
@WorldEngineer - Yes, but there are an infinite number of curtains
 
user20683
2:58 AM
@psr Zeno and his theatrics...
 
Though I've heard on multiple occasions people say the UNP's not a very good book. I found it illuminating enough but I was 15 so my memory's may betray an illusion to me
 
user20683
@JimmyHoffa I suppose I could always read the networking book that my school uses
 
user20683
it's pretty much used everywhere
 
user20683
There's also the above option if I want to wait a tad
 
3:00 AM
@WorldEngineer from a distance washing dishes I turned around to read that on my screen "if I want to try a fad", quite apt...
 
user20683
@JimmyHoffa yeah
 
user20683
I don't consider online courses to be that good.
 
I'm withholding judgement until I've tried; I learned everything I know online after all (by and large anyhow), but I'm also not holding my breath..
 
user20683
@JimmyHoffa I made it partially through Stanford's trial run AI class
 
user20683
I needed to have had more math before I tried that
 
3:05 AM
I've long held the belief there's no such thing as a teacher; logically someone can learn without a teacher, and someone can with a teacher not learn, so the teacher is totally ancillary to the learner. Though it's more complex than that, people learn in such different ways from each other... many may not learn without a teacher I suspect.
 
user20683
@JimmyHoffa depends on the teacher
 
user20683
in my experience
 
user20683
some teachers I've learned a great deal from
 
user20683
most, sadly, not so much
 
3:19 AM
Aye, either way I don't suspect the online learning technique fits with the learning style of a large portion of folks regardless of the quality of the teachers or course content, but time will tell if I'm right (or hopefully, dead wrong, how nice that would be for the world!)
What did you think of stanford's AI course?
Also, you didn't answer me; what are all the HTTP verbs you know off the top of your head?
Is web dev what you want to do? Thought you had said you were interested in embedded at one point
 
user20683
3:39 AM
@JimmyHoffa Flexible but headed to bed
 
user20683
sorry realized my phone/alarm was not where I thought it was
 
user20683
so had to look
 
8:57 AM
I'm facing an issue with the translation of a project I am contributing to. The problem is about how to handle translation of hotkeys. Would that be a question for Programmers?
@OliverSalzburg Seems like I found a home for it on User Experience :)
 
-1
Q: PhoneGap and JQuery

amira Rashed I want to develop an application using eclipse and PhoneGap and JQuery but i get a validation error when i create a phonegap project and include JQuery library . Can you please help with that error ? Thanks

I am always tempted to suggest reinstalling Windows if people can't include the actual error message.
2
 
 
4 hours later…
user41796
1:19 PM
are 10k+ users supposed to be able to search for deleted questions? I remember being able to do so at one point in time, but the "deleted:yes" search parameter doesn't appear to be working now.
 
@GlenH7 No. If deleted:yes ever worked for you, that would have been a bug.
Perhaps this is what you remember?
10
Q: Search brings up deleted questions in results

Yannis RizosCouple of days ago I started noticing deleted questions appearing in search results. At first I thought it was mod vision (moderators can search for deleted posts) but today when I happened to search in porn incognito mode: The circled question10K+ link was deleted on February 9 and it really ...

 
user41796
@YannisRizos - yes, that's what I used to verify I could at one point see deleted posts. I must have misunderstood your comment at the time.
 
user41796
That means the one time I was able to see them was due to the bad shard that Nick explained.
 
user41796
Bummer. Since there's a question about context driven testing related to IEEE 829 that would be of some relevance to someone asking about CDT now. But I think it may be the same user.
 
Yes. And it wasn't just 10K+ users, deleted questions were showing up on search results even for non registered users.
@GlenH7 Is this what you are looking for? programmers.stackexchange.com/questions/197957/…
 
user41796
1:34 PM
that would be bad. :-) I still wish I could search on deleted Q's
 
user41796
ding! that's the one. And it's the same user. Which resolves to "don't bother" on the new question
 
user41796
Not sure why he deleted his question. It's kind of an obscure area so it may take a while to see activity. OTOH, the whole CDT vs IEEE 829 issue is apparently a bit of a flame war in the testing circles
 
user41796
1:55 PM
@YannisRizos - do you know if requests have been made on MSO to be able to search for deleted questions? I have searched but can't find anything directly related
 
user41796
thanks. I had a feeling it had been discussed and received a cool, if not frosty, reception
 
I can't find a feature request either, but... it's not going to happen. Kinda defeats the purpose of deleting stuff.
40
Q: Enable searching with deleted:1 for 10k users

bfavarettoI just tried to find a recent answer I wrote (or comment, can't remember), but I couldn't. I assume the question was deleted. Can the deleted:1 operator be enabled to 10k users? We can see deleted posts, so why can't we at least target them in search? I understand that allowing us to search for ...

@GlenH7 Not declined, but... ignored.
Also, apparently I thought it would be a good idea at the time.
 
user41796
I hear ya
 
user41796
it's nice for sleuthing
 
user41796
2:02 PM
which is perhaps a sign that I should be doing so... :-)
 
user41796
"which is perhaps a sign that I should not be doing so"
 
user41796
This one is more directly related; and has an answer: meta.stackoverflow.com/q/41130/194162
 
user41796
And it's officially declined. Not sure that I agree with the reasoning. The "against" arguments are even more rare than the in-favor arguments
 
2:26 PM
is it accurate to say with 3 bytes for RGB, the furthest distance in the color spectrum from one color, is that color + half the value of ffffff ?
No, that doesn't make sense..
 
none at all. what would be a distance in the color spectrum? the furthest distance being from red towards blue or purple?
mapping rgb to "real" colors would be a pain as everybody can tell you who ever tried to mix colors in rgb for image processing.
 
Yeah..
1
A: What algorithms are there for picking colors for plot lines on graphs?

Jimmy HoffaTry and pick a visible range of brightness, say #444444 - #ffffff or whatever it may be, then treat the whole number (r/g/b) as a single value, divide it by 12 and pick your 12 colors as even spaces inside your visible brightness range and put them in an array. Now you have an array of color val...

Here's my algorithm anyway that i came up with.. maybe it's not that particularly good
I'm imagining those color wheel spectrum thingies, if you go more than 180 degrees away from your current location, you've gone closer to where you are, you must go no further than halfway around to get the maximum distance away from your current location
but #RGB isn't an appropriate mapping to those color wheels..
 
That's more a visual problem. The colors he uses in his first example are mostly as far from each other as you can go if you need twelve different and avoid too dark colors.
His second example has only five graphs, so not too difficult to make it work.
Your approach to get as much distance as possible (plus avoiding the darker colors) would be mostly the best you can get. From there on he should rethink his UI and maybe just avoid having too many graphs at the same time.
 
user41796
2:42 PM
Any thoughts on why Mark Trapp deleted this answer? It's essentially a link-only answer but it does appear relevant to the question.
 
@GlenH7 per my reading, deleted answer rather blatantly violates mod notice posted in the question: "We're looking for long answers that provide some explanation and context. Don't just give a one-line answer: please explain why you're recommending it as a solution. Answers that don't explain anything will be deleted..."
> The book answer many of your questions :) 37signals.com/rework It changed the way I think about creating software and building companies.
 
user41796
true although the notice was put up several months after that particular answer. I'm not saying it was a great or even all that good answer. Just not necessarily worthy of deletion
 
one liner that lacks quality explanation. Without a notice, deletion would be somewhat slippery - by default, it's not too criminal. Notice makes a difference though, at least to me...
8
A: How can we encourage down-voting over deletion on answers?

gnat"Rules of the game" have been set just few hours after the question has been asked. Post notice in the question says clearly and unambiguously: We're looking for long answers that provide some explanation and context. Don't just give a one-line answer: please explain why you're recommending ...

@GlenH7 time wise, notice and deletion appear pretty close don't they?
 
user41796
that's fair regarding the notice. It's a pretty subjective question that has attracted some crap answers. Mark was probably trying to be thorough in keeping the question clean
 
> Notice added Insufficient explanation by
occurred Jan 21 '12 at 3:20
> deleted by Mark Trapp Jan 21 '12 at 3:17
 
user41796
2:49 PM
yes
 
user41796
looks like it hit his radar for whatever reason and he went through cleaning it up. Then explained why it was cleaned up via notice
 
@GlenH7 good guess, this seems like the most likely way things went back then
 
user41796
and sometimes now too. ;-) I'd taunt Yannis about being an evil-mod but he already dropped off
 
> Without above notice, I would rather prefer downvoting bad answers instead of their removal. I downvote quite frequently and rep points I loose for doing this feel like an "investment" into site quality; pretty rewarding. But wasting reputation on those who just can't play by rules... give me a break.
:)
 
@thorstenmüller Having actually done this a long time in the past - the color distance in the "Hue" component of HSV and HSL is much more useful than poking around in RGB hoping for a good perceptual color distance. I added that bit to my answer, since I didn't originally motivate why I'd pick one of those color spaces over RGB.
 
user41796
2:55 PM
if nothing else, I was reacquainted with 37signals, which had dropped off my radar
 
user55340
@JimmyHoffa Thats why you switch to HSV.
 
user55340
 
@MadKeithV Why do you pick an arbitrary distance instead of a maximum possible distance though?
@MichaelT Yeah, I realized that after reading @MadKeithV's answer that the color wheel I was trying to picture was HSV/L
 
user55340
I would inscribe a polygon of the appropriate number of sides inside the HSV at a particular V. When you run out of colors at that n-gon, you then move to another value of V and repeat.
 
user55340
Its an easier problem than trying to get a maximal distance 3d structure inscribed in an rgb color cube.
 
3:05 PM
@MichaelT that makes perfect sense, I still think my approach to rotating X number of colors is ideal just rotating H like you said as opposed to just picking an arbitrary distance
@MichaelT

(defn hueColors [x] (map #(* (/ 255 x) (if (even %1) (+ %1 (/ x 2)) %1)) (range 1 x)))
there's your hue selection, N choices with each one maximal distance from the other
(almost anyway, I didn't handle roll past 255 correctly, whatever just needs a mod 255
or mod x rather
you can then rotate that list and each time you hit the end increase your v, does that make sense?
 
user55340
Neat bit - S and V are floats between 0 and 1, and H is either 0-359 or 0-1.
 
@MichaelT does my approach make sense?
 
@JimmyHoffa Maximum possible distance depends on your graph coloring complexity - and that gets hard fast. It's easier/quicker to pick a distance that's "different enough" and keep the algorithm simple until you NEED it to be more complex.
 
user55340
The other thing is... if there is a certain maximum number you can deal with... just create that set and load it in rather than calculate.
 
@MadKeithV I understand the graph coloring problem is significantly more complex, but in this case each node only has 2 edges so it becomes an arbitrarily simple alternation
 
user55340
3:18 PM
I doubt if Excel or Numbers calculates "red first, then blue, then green" when doing the colors for the graphs.
 
Unless you're paying attention to graphs that have lots of crossing and interplay between the bars, then the coloring problem becomes intensely more complex and you're right, he needs to then study the graph coloring problem if he cares
 
user55340
It costs very little to have an array of 128 (or 256) color objects. Eaisly changed for themes too...
 
Aye
 
user55340
For example you can 'theme' it for greyscale, or high contrast, or color blind...
 
Yes, a fixed color list is really the appropriate implementation, but the algorithm is still a fun puzzle :) So long as the graph only has two edges for each node, it's super easy
 
user55340
3:23 PM
Oh indeed... I've thought of it too and it is a fun one.
 
My light table is being a hoebag
@MichaelT does that clojure work in your light table?
 
user55340
I haven't poked it yet... give me a bit... trying to find this old site that I remeber that was just people submitting color themes.
 
user55340
Sometimes it was just 3 colors, othertimes it was a dozen to be used sequentially.
 
Ah even isn't a function, had to define it
 
user55340
Something like thus...
 
user55340
3:26 PM
 
user55340
colorschemedesigner.com is fun to play with..
 
@MichaelT That reminds me of a site I found the other day for choosing color themes for UI that was interactive... was really neat...
Oh, that's it :)
 
user55340
There's also kuler.adobe.com - but I don't have modern flash on this machine.
 
user55340
And colourlovers.com - in particular the pallets section colourlovers.com/palettes
 
(128/3 256/3 128N 512/3 640/3)
Wth kind of numbers are these ending with n?
 
user55340
3:32 PM
All Numbers start with N. So do the Natural ones...
 
let's see, haskell would do the same algorithm as...
colorSlice n size
  | odd n = ((sliceCount / 2) + n) * size
  | otherwise = n * size
  where sliceCount = 256 / size

hueColors n = [colorSlice x (256 / n) | x <- [0..n]]
 
 
4 hours later…
user55340
8:01 PM
@WorldEngineer did that "how to build a jar file" fail a migration?
 
8:17 PM
@MichaelT It failed for two decades if you ask me...
 
user55340
I just wondering... I've got a comment on there and the question was closed before I was able to toss my own close-migrate vote in there.
 
waiting for some horrible sql query to finish which is going against a fairly large amount of data in a database with zero quality hardware
wee...
 
user55340
(whee! its fun to see a +15 rep gain)
 
8:33 PM
go query go... finish...fin...agh...
18 minutes and counting...woo... this database will eat it's words when I'm done with it :|
...22 minutes...tell me it will finish, please please tell me it will finish...
huzzah, I win!! it completed! ARRR!!
 
 
1 hour later…
9:44 PM
Alex Miller on May 14, 2013

We’re Back!  It’s been a while since our last podcast (why you ask – listen to find out!) but we’re back now and “stronger” than ever.  It’s Joel, David and Jay (plus producer Alex and Abby) coming to you from the brand new SE Podcast Studio (check out the picture below)

News of the day: we’re finally in our new office (and podcast studio). We’ve got hexagonal offices (and therefore crooked hallways), and a cool café area. AND HEATED TOILET SEATS. And a kitchen with a giant walk-in refrigerator, for our interns (which we don’t have). …

 
10:38 PM
0
Q: Document existing Visual Studio Project

Walt DietzI am a small business owner and my sole developer quit. Cold. I am not a coder. He developed an existing application that works .net, c#, javascript, sql, visual studio based, 250,000 lines of code. Needs to evolve to v2.0. I have zero documentation. Is there a product that can read the code and...

I really want to just comment "...you are so screwed..."
Nah, if the dude was a good coder (which if he wrote a 250kloc app on his own either he's great, or abundantly awful) then another team will be able to pick apart the remains and get acquainted with it on their own (unless they're also terrible)
 
11:06 PM
Hello and Welcome to Programmers. Please don't post questions here to circumvent question bans on Stack Overflow. Have a pleasant day. — World Engineer 4 hours ago
despite the mod request, they continue to trash the page with more stuff like that
sounds like a clear case for deleting their questions (all but the first one) and account suspension for a day or two to cool off
 
user55340
11:35 PM
@gnat Given past experiences, I believe that 3 out of 3 questions downvoted and closed will result in a question ban here too. I believe a suspension will be unnecessary unless one is really trying to make a point - it won't change what he can do much at all.
 

« first day (984 days earlier)      last day (4005 days later) »