« first day (1321 days earlier)      last day (3663 days later) » 
00:00 - 19:0019:00 - 00:00

7:00 PM
@AJHenderson Zoom.... Enhance
 
user55340
@RobertHarvey 380 questions, of which when you hit the bottom of the first page (13 pages) its a score of +8.
 
user55340
Median question score is +1.
 
user55340
And 4 pages (of 13) of +0. I'm not suprised at all.
 
@Ampt yeah, pretty much
 
user55340
 
user55340
7:07 PM
Gotta go with the clasics.
 
@MichaelT $38000?!?! You could get a late-model used corvette for that price
 
user55340
@Ampt That was back then, not sure what its now... then there's also the problem of "how quickly can you store a 160 megapixel image to some media"
 
user41796
@MichaelT Sigh. I hate systems where you have to let it fail before you can get a reaction.
 
@MichaelT I think my friend had something to say about that. What's that J?
 
user55340
The price back then was a fair bit of "you need a gigabyte of ram in the camera" so that you could capture the image to something thats fast to write to...
 
user55340
7:11 PM
and then you typically had a laptop connected to it via fiberchannel or similar so you could write the data from ram back down to some other media.
 
And here I thought the 60D I was looking at was fancy
 
user55340
That said, its a scanning back... or it would be even more expensive.
 
user55340
 
user55340
And the 6x17 is about 7" long.
 
user55340
For telescopes they sometimes use a 2xN sensor array.
 
7:14 PM
wonder what kind of sensors they put on those spy satellites haha
 
user55340
Smallish (compared to that) with the image used over the area at the time.
 
user55340
 
wonder if each of those panels can be manipulated independently to get them as aligned as possible
 
user55340
They're closely aligned, but not manipulated after mounting.
 
really? Surprising
 
wonder what it'll take to get edgeless sensors
 
user55340
But this is only really practical for telescopes and such - not regular cameras. Each chip has its own profile.
 
user55340
@Ampt How do you read the data out?
 
bottom?
(admittedly I'm woefully ignorant of how that process would actually happen)
 
user55340
Not as easy and rather expensive.
 
user55340
7:19 PM
CCD has the nice 'march the charge down to the edge' approach.
 
well, I imagine the fuel used to park it in orbit is also expensive
@MichaelT Really? That's my major.
 
user55340
@Ampt You won't find things such as "testing" and "builds" and "version control" as a topic in a CS class.
 
@MichaelT OH
I guess I didn't understand the line of topic as well as I thought
 
user55340
And frankly, most SE classes I've seen students come out of aren't much better...
 
user55340
Colleges are typically teaching students how to learn and how to be academics - rather than how to work in the industry.
 
user55340
7:23 PM
If they learn the how to learn well, its not an issue.
 
user55340
But you get a lot of students who go from college to industry and go "this isn't anything like what we did in college" - the "you maintain the software forever rather than until the next project"
 
@JimmyHoffa hard to tell, these sound quite original. Remind of Sex Pistols, but not very close. I think I heard something like that before but can't recall. Try Clash maybe...
 
Sometimes I wonder if the only experts left on Programmers are also the same ones who frequent this room.
 
user41796
@RobertHarvey it does seem cyclical
 
@MichaelT I have to say, the SE course I visited (which is mandatory in my Uni's CS curriculum) covered version control with SVN (it has switched to Git since then) and testing in various aspects (coverage metrics, unit test, …). It was however academic enough not to focus on something like jUnit.
 
7:30 PM
@amon Interesting. I had a whole course focused on testing in which we played with jUnit, code coverage tools, etc.
granted it was using JU3 which was about as outdated as you could get at the time
 
user55340
From what I see from interns / new hires in the past years is that they don't necessarily understand the "why" of it - its either a dogmatic religion or a "why bother, it didn't help us at all in college"
 
user55340
And if you get to things like CI and static analysis or code reviews...
 
user55340
Lots of fast and lose cowboy coders out of college - not so much the disciplined engineer.
 
user41796
@RobertHarvey - I've been pondering some of the meta-meaning behind site participation lately. One somewhat interesting anecdote from the conference that I am, is that I ran across "Ross" over on SO. Of course, he didn't say _which one_, so it makes it a bit more difficult to recognize folk.
The broader pondering is what are the rewards of participation beyond unicorn poo? What external recognition does it provide or should it really have any as that can indirectly encourage more gaming of the system.
 
user55340
But you've also got things like "here's code written by a grad student 2 years ago.. this semester you're going to be adding new functionality to it..." which would be a good class.
 
7:34 PM
You only learn the discipline on code projects longer than three weeks.
 
@RobertHarvey programmers.stackexchange.com/… quarterly top league seems to prove you wrong. This room regulars rather loose rep, wasting time here instead of rep-whoring out there in the wild :)
 
Or from actually suffering frm not having the discipline
 
user55340
@Zeroth Even 3 months is a bit short.
 
I learned just how important source control was after I worked with an engineer and a math guy in AI on our class project
the engineer didn't want to bother with source control
 
user55340
@gnat we are high up in the leagues though.
 
7:35 PM
math guy disappeared with our code, got sick and was gone for three weeks
then he lost our code
so we had three days to finish all required functionality of the project from a 2-months outdated archived copy of the code I managed to find
 
@MichaelT you and Robert, yes. Sort of exceptions that prove the rule
 
user55340
 
user55340
You and Glen are up on that chart reasonably too.
 
user55340
A reasonable number of people on that list show up in chat on a not irregular basis.
 
user41796
@MichaelT Wow, I've taken a nose dive in my quarterly rankings. :-(
 
user41796
7:39 PM
Soon enough things will be calmer
 
user55340
Stress does that - remeber that graph of the last months at my former employer?
 
user55340
 
user41796
yep
 
user55340
The thing is also there's a few different categories for the 'expert'. You've got 'active' vs 'less active' and 'expert' vs 'not expert'. There are quite a few less active experts on the site... but the most active experts tend to be involved in the community too - and that likely means here.
 
69
A: What aspects of psychology does Stack Overflow take advantage of?

AarobotActually, the main psychological phenomenon underlying Stack Overflow (and numerous other sites with similar reputation/karma/badge/reward systems) is something that's been discussed on meta before: Intermittent Variable Reward. Although I think the "correct" term (or at least the one more commo...

 
7:44 PM
@gnat Do any of those people use their close votes?
 
user55340
Some of the less active experts are, however, active elsewhere on SE (or other communities).
 
user55340
Very few people comparably are active in close votes... and even fewer are active in /review which gets the old ones.
 
user55340
The only close votes I see from less active people are on egregious things, or things that get flagged low quality and show up in the blue box notification.
 
@RobertHarvey yes, I recognise a couple of frequent CV reviewers among them. CV queue isn't really chat-driven (which should be a good thing, I hate gang voting)
 
user55340
At times I wonder if SE tries leaning on the community a bit too hard expecting them to do the moderation without giving enough incentive for it, or stronger social aspects to help form a stronger core community.
 
user55340
7:47 PM
And then you've also got the... for lack of a better phrase... inclusionist 'faction' that doesn't downvote or close vote for some reason (not hurting feelings or the like).
 
user55340
(btw, Eric Lippert did three delete votes yesterday - I noticed because I was the third vote on each)
 
@MichaelT Different philosophies.
 
user55340
@Ampt Yep. But the thing that I tend to see is the more active you get - reading the front page every day and looking at every question the less tolerant you get for the stuff that bumps the good material off the front page.
 
@MichaelT they are lightening recently, with tricks around hot questions making these not that hard to moderate. On the other hand, their recent experiments in LQ queue (quite tectonic I think) seem to be directed on forming wider moderating community, though this is mostly pronounced at SO (and maybe SF/SU)
in a day or two I am going to post at TWP meta some observations on that
 
@gnat sex pistols... yeah that is a good instinct actually, it's easily been 10+ years since I last listened to them, perhaps I should toss some of that in
 
user41796
7:51 PM
@gnat I think the shifts will show sooner on the big 3 because of their volume. Smaller sites will (naturally?) take more time to see those changes take effect.
 
user41796
@MichaelT I think that eloquently captures a chunk of my recent musings
 
@JimmyHoffa pure luck indeed - I just happened to hear something like that played by pistols 3 or 4 weeks ago :)
 
user55340
SE tries very hard to not be a social site... which makes it that much harder to form the core group communities they need to do the moderation.
 
@GlenH7 yes, I also think so. Another thing is, at smaller sites it's technically possible for moderators to give more attention to review queues, which fuzzies features intended for larger-scale
 
There are downsides to being a social site
 
user55340
7:54 PM
@Zeroth Yes... read shirky.com/writings/group_enemy.html though
 
user41796
@Zeroth agreed; which is why SE avoids turning in those directions.
 
Oh I've read them. The downsides are that you end up with cliques and divisions. It seems that divisions in a socially-focused site are almost bound to happen.
 
user55340
You've read the Shirky essay?
 
Probably because of geek fallacy #4, friendship is not transitive
I have several of his books
He published a third recently I need to pick up I think
 
user55340
Did you read that specific one? In particular the 3 things to accept #2.
 
7:56 PM
@Zeroth Does that essentially mean "A friend of a friend is not necessarily your friend?"
 
I remember reading it but not specific details...
Yes, Robert.
 
user55340
The issue is that the 'participation on the site' (measured by rep) is the measure for 'part of the group that can moderate'... however, most of the don't - the core group does.
 
Thanks, just checking.
Had to look up my basic math.
 
Oh man, transitive is such a good word though
 
user41796
@RobertHarvey 0.1f + 0.2f != 0.3f
 
7:57 PM
Ooh nice example.
 
@MichaelT this is all hard to think about though. A problem? maybe. Solvable? unlikely... It's only a problem if you can make some distinction of threshold that other's show greater success than us and call that the threshold of problem vs. not problem. I don't know of other sites more succesful than SE in this goal, the only one that comes to mind is wikipedia- I'd say they have greater success in their community-content-creation endeavor than SE. So here's a question: What do they do
to engage and encourage community moderation?
 
user55340
@JimmyHoffa Possible, though I'd almost say "fork the site" - you've got meta, chat, and the main site... what about "discourse.programmers.SE" or "social.programmers.SE"?
 
user55340
Where discussions can happen on discourse and a google+ like structure on social.
 
user55340
@JimmyHoffa And the answer is very little - you get a badge for doing reviews... and a hat sometimes for participating in closing a question.
 
@MichaelT Discourse requires new software. Most of the value of the SE software lies in the cultural expectations that are encoded within it.
 
user55340
7:59 PM
Its that gamificiation that drives a lot of the activity.
 
user55340
@RobertHarvey a discourse like structure... something where a discussion could be had in non-real time. Formesque, but better.
 
@MichaelT wikipedia has questions?
 
user55340
Let there be another site that just shares the same user profile as the other site... and thats it.
 
While wikipedia could be argued to be very successful, it ends up with a lot of the pitfalls SE seems to want to avoid
you get protectionist editors, bike shed debates all the time
 
user55340
@JimmyHoffa Wikipedia doesn't have gamification, but it does allow communities and discussions to form more eaislly.
 
8:01 PM
@MichaelT I hope Jeff Atwood can figure that out. I like some of the material I see on Quora and Redditt, but really dislike some aspects of those platforms.
 
@MichaelT how ? That's what I was asking - what do they do that has the result you refer to?
 
user55340
@RobertHarvey I'd love to see the discuss.P.SE with enough interlinking so a question could be migrated 'sideways' to that site.
 
I've no familiarity with wikipedia other than as a passive visitor really
 
Quora, for example... Once you get involved in a question, it sends you rampant emails about the activity on those questions. I don't care if someone else posted a new answer, unless I'm following a question. You can't turn it off.
 
user55340
@JimmyHoffa mailing lists, messages to a person on their talk page...
 
8:03 PM
@MichaelT mailing lists... that's an interesting concept; how could that work with SE I wonder?
 
user55340
Not shutting down discussions quickly. Allowing the community pump.
 
10
A: Replace accept rate with citizenship level

gnatIt would be interesting to give it a try by showing "citizenship level" value at MSO for 2-3 months. As for how to calculate the citizenship level, I am pretty positive that whatever would be measured, it would better be expressed in logarithmic scale: it would be much less interesting to know i...

 
@MichaelT that would be bad here though.
 
@JimmyHoffa You haven't seen pedantic until you've seen some of the discussions they have there over seemingly trivial facets of an article, and whether or not they're "encyclopedic."
 
@RobertHarvey I think the key is wikipedia keeps those things off the main
 
user55340
8:04 PM
@JimmyHoffa Here, Programmers.SE - yes. Here "community.Programmers.SE" hypotehtical site... maybe not.
 
@JimmyHoffa That's a big part of it, yes. But my God, the pedantry. Makes the C++ folks look sloppy.
 
maybe if SE allowed those types of discussions in a fashion that kept it out of the passive googlers eyes....
@RobertHarvey as to your last point there; impossible.
 
user55340
You ask a poll question? Lets migrate it to discuss.P.SE rather than closing it as too broad.
 
@JimmyHoffa You apparently haven't stumbled over one of the aforementioned discussions.
 
user55340
You want to follow a person's "advertised questions and answers" or blog posts or private message them? google+.programmers.SE or whatever.
 
8:06 PM
@MichaelT FB with Q/A instead of arbitrary posts, no pictures or videos etc... there's a weird thought :o
 
user55340
The thing is to allow a stronger community to form around the mini-network of sites of programmers.SE that would allow for a larger core group interested in the moderation of the sites.
 
While the C++ folks can be extremely... precise
 
user55340
@JimmyHoffa its called "linked in"
 
the wiki discussions get very very heated over very minor things
let me find an example
 
@MichaelT linkedin has Q/A?
 
8:07 PM
@MichaelT Tag Communities would be one way to do this. There's so much potential in Tag Communities that's unrealized.
 
user55340
Not so much Q/A but rather 'advertise what you wrote to potential employers'
 
@MichaelT not the same. I'm thinking more like, what if SE allowed you to friend people and get an aggregate feed of all those peoples Q/A... (maybe they do already?)
 
user55340
@RobertHarvey Yep. Default 'hangouts' or similar built on each tag... could be interesting.
 
user55340
@JimmyHoffa They don't and they try to avoid it. Its been suggested on MS[OE] several times.
 
user55340
5
Q: Follow a person?

IAdapter Possible Duplicate: A “Friends List” on StackOverflow would be nice I'd like to have feature like Twitter (I think; I don't use it) to follow a person and be updated on new posts by that person. I do know that I can just click on that person's profile once every week or day, but I thi...

 
8:09 PM
@MichaelT I'm just curious as a thought experiment what the result would be
 
wikidrama.blogspot.ca <-- good set of examples here
 
The trick is getting people to do the work. "Oh, just write something for the blog." Hm?
 
@RobertHarvey if they had followers it might encourage that
 
user55340
yesterday, by MichaelT
Btw, for prospective P.SE blog authors: https://stackedit.io/viewer#!provider=gist&gistId=e3e1df880399a55a3780&filename=‌​progsesilentvotes.md could be useful for topic choices.
 
knowing that a group of people whose opinion you value will automatically get the content posted to their feed when they come to the site
curious thought
 
8:10 PM
@JimmyHoffa It might. But if I want to see what Eric Lippert or Jon Skeet is up to, I just lookup their accounts.
@MichaelT Is that as simple as providing an RSS feed for accounts?
Or is that not simple?
 
user55340
@RobertHarvey Could be... or slurping the network profile page.
 
@MichaelT technically this could be done with a clientside scriptlet, but would never spread to have any community-wide effect while it required third party client side extensions
 
@JimmyHoffa There are a couple of things like this available on Stack Apps, but it seems pretty pointless to me. I discover content by topic or technology, not by author.
 
user55340
@amon There are some authors you want to say "oh, what did they say" - when Eric shows up on P.SE, I do want to read it.
 
user55340
(And frankly, when I see that 'modified by red-orange mostly square' in question activity, I click it too... word recognition...)
 
8:34 PM
Well, thank you :) Of course, I have more trust in certain known users than others, which translates in speedier upvotes. But I personally don't see the point of reading something well-written that has no relevance for me (say, a nice and eloquent answer on some C# specifics).
 
user55340
Arguably, when Eric shows up on P.SE the question is rarely about the specifics about C# but rather some larger language design issues and sometimes how C# applies to them.
 
user55340
(only 32 of his 78 answers are even tagged C#)
 
8:55 PM
That, and his knowledge is deep enough that, even if you don't know the language he is writing about, you can still glean insights. His Hotel Room metaphor for invalid memory accesses is classic.
 
user55340
He's one of those "when I read his answers about design theory I have to think - and thats a good thing"
 
Would it be possible for me to ask one of you a question that no one on SO has yet to answer? It's in preparation for an exam, and I can't figure it out
 
Don't ask if you can ask, just ask.
It's a chat room. Your question will either get answered, or it will get ignored. :)
 
Ask if you can ask, @RobertHarvey's just a grumpy old crank. I appreciate how the pointless noise annoys him and others.
 
if you ask to ask make sure you have a counter or something to avoid stackoverflow
 
9:00 PM
0
Q: Assembly language - third element points to

ShahrukhKhanwould really appreciate if someone can tell me what the third element in the strong list reads. This is NOT HW, I am merely preparing myself. Thank you.

 
@JimmyHoffa That's Mr. Grumpy Old Crank to you.
 
Robert Harvey has 282K rep, he is famous
 
@ShahrukhKhan It reads 0x00001000 0x00238480.
 
How did you figure that out?
How did you get that answer, Im sure you figured it out because you have experience
 
[sigh]
When you're dealing with hexadecimal numbers, each group of two digits is a byte.
@ShahrukhKhan I take it 0x00238480 is the memory address of the string?
 
9:05 PM
yes
what does the third element in the string list read, is the question
it's a linked list with three elements
 
head is at 0x00001000 so the pointer for the next string is at the next cell which reads 0x0030000 and so one more and you reach 0x00000010
 
@ratchetfreak Isn't that the second element, not the third?
 
it's the third if you count *head as an element, I went through 3 nodes
 
Right, OK. So the address to the third string is 0x4024FFA4
 
An explanation would be sincerely appreciated for a noob like myself.
 
9:15 PM
And the data at that memory address is 0x504D4F43.
@ShahrukhKhan: Well, the first thing you need to know is that memory addresses increase from the bottom to the top of the diagram.
 
understood
 
The second thing you need to know is that a pointer is just a pointer to another memory address.
 
ok understood...
 
Each element in the StringList consists of two memory addresses... a pointer to the string, and a pointer to the next element in the StringList.
 
understood..
 
9:18 PM
You start at memory address 0x00001000. That's where the first StringList element is located.
 
understood
 
0x00238480 is the memory address of the string pointed to by that StringList element.
0x00030000 is the memory address of the next StringList element.
 
understood
 
The next StringList element is located at memory address 0x00030000. The memory address of its string is 0x00008000, and the memory address of the next StringList element is 0x00000010.
 
understood
 
9:22 PM
Lather, rinse, repeat. Until you get to the third StringList element.
 
But now he has 3 values
1 for his memory value
but why are there 2 more
that is my confusion
 
You mean at 0x00000010?
 
yes, so now this.next is the 3rd element
 
to trick you up
no it remains the second
 
0x4024FFA4 that's the memory value of the string
memory address of the value of the string of the 2nd element
 
9:24 PM
If you're referring to the third memory space above 0x00000010, that's a "don't care."
 
but we are concerned with the 3rd node, which now all of a sudden has two condenders
 
so now I go to *4100
I did not know that it's considered a dont care
You taught me something
 
9:25 PM
Well, wait.
 
whose value of the string is *C84C
 
You go from 0x00001000 to 0x00008000.
Then from 0x00008000 to 0x00000010.
Then from 0x00000010 to 0x00041000.
The node at 0x00041000 is the third node.
That's where the textbook question gets a bit muddy. It says "What does the third element in the String List read?" I interpret that as "Decode the string located at memory address 0x0A00C84C.
 
I get it
Wow
Robert you are awesome
 
No, just been doing this for far too long.
 
ratchet freak so are ou
What do you work as?
 
9:30 PM
I'm a programmer as it happens...
well sometimes I also regulate traffic lights
 
ratchet freak is the reason we wait so long at red lights.
I knew it was somebody's fault.
 
no I'm the reason you see buses get all greens while the rest waits
eitherway I'm going to take a shower
 
Robert Harvey are you a Software Developer?
 
Yes.
Technically Software Engineer. But what's in a title?
 
letters
 
9:34 PM
Exactly.
 
sometimes numbers
 
Nice
 
0
Q: Confusion on Kendo UI changing licence from GPLv3 to Apache 2.0

Ankan-ZerobWhat are the problems and reasons for Kendo UI to change their licencing from GPLv3 to Apache v2.0. The reasons cited in their FAQ, according to my understanding of GPL v3 do not seem accurate, Can the community please clarify this: from : http://www.telerik.com/kendo-ui/core-faq but because...

I smell an agenda.
 
user55340
They likely miss-worded the reason - instead of "can only be used for non-commercial projects" it should likely read "can only be used for non-proprietary projects". That said, the best people to ask about this are telerik themselves. — MichaelT 1 min ago
 
@MichaelT Yeah, but what in the GPL prevents Telerik from accepting customer contributions?
 
user55340
9:40 PM
There's some less appealing patent things in the gpl v3.
 
Yeah, that would do it.
 
user55340
> You may not convey a covered work if you are a party to an arrangement with a third party that is in the business of distributing software, under which you make payment to the third party based on the extent of your activity of conveying the work, and under which the third party grants, to any of the parties who would receive the covered work from you, a discriminatory patent license
 
[sigh] My head hurts already, just from reading that one paragraph.
 
user55340
Paragraph? I don't see ay periods in there - thats one sentence (and there's more after it too)
 
I don't need to see it. :)
 
user55340
9:43 PM
Section 11 - plain english: gplv3.fsf.org/dd3-faq
 
user55340
> Second, in the seventh paragraph, the draft says that you are prohibited from distributing software under GPLv3 if you make an agreement like the Microsoft-Novell deal in the future. This will prevent other distributors from trying to make other deals like it.
 
So basically, if you're in the business of enforcing your patents, the GPL is pretty much incompatible with your company.
 
...really?!
 
user55340
Yep. And if any of your employes contribute as part of the company work, there could be a mess.
 
No wonder v3 scares so many
Because that doesn't seem to exclude non-software patents
which I could see an engineering company holding and enforcing
 
user55340
9:45 PM
Its one of those "do you really want to try enforcing this in court?" battles that if you lose you lose in a big way... so just don't touch GPL at all.
 
user55340
I'm BSD type... though sometimes I'm tempted to look at the MicroSoft Public License.
 
In other news, Kendo Core going permissive is way cool.
I never thought the GPL was a good idea for Kendo.
 
user55340
The other thing it does for them, is allows them to make some bits of their code commercial... and well, they get paid.
 
user55340
>
Not all Kendo UI Widgets are created equal and some, like the Grid, Scheduler and Editor, consume a great deal of engineering effort to build and maintain. Widgets such as these are complex, but they are also essential for customers building certain types of apps, and the open-source equivalents for these leave a lot to be desired. To ensure that critical widgets like the Grid, Scheduler and Editor continue to get the proper attention and focus they deserve from our engineering team, we’ve decided to offer these only with commercial Kendo UI licenses.
 
user55340
With an apache license its much easier to let the two parts mix.
 
9:51 PM
I can live with that. It's a much better arrangement than the GPL/Commercial schism.
MSO/MSE split will begin momentarily. Say a prayer.
 
10:07 PM
 
@RobertHarvey technically it's the bureaucrat that can't help but decide what color the bikeshed should be
 
 
1 hour later…
11:22 PM
If somebody was almost done assembling a car when their boss came over and says "oh hey, the customers would like us to attach a jet engine to it, can you have it done in a couple days?", cars would be pretty unreliable too. — Anna Lear ♦ Jan 28 '11 at 22:23
 
00:00 - 19:0019:00 - 00:00

« first day (1321 days earlier)      last day (3663 days later) »