« first day (1291 days earlier)      last day (3693 days later) » 
01:00 - 20:0020:00 - 23:00

1:55 AM
@RobertHarvey funny, I thought you were doing a pretty spot on job of harvesting those precious stars :P
Almost as valuable as that elusive MSO rep.
 
user55340
2:11 AM
I've come to realize that this question
 
user55340
-1
Q: How to create a some kind of value for sentences?

user3149I want to identify most matching sentence using some pattern. That means by using java algorithm I want to create identical value for each sentences.Each sentence when entering to that algorithm can be out some kind of identical value. How can I develop it? What can I refer any web sites do you k...

 
user55340
Is someone trying to build a translation tool based on an enumerable number.
 
4:50 AM
oh god I never counted on having to see people's flags
I just wanted to laugh at the silly deleted content, but seeing the flag queue makes me want to bounty myself back below 10k
2
 
 
2 hours later…
6:34 AM
@JimmyHoffa simplest way to get rid of this headache is not to look at flag queue at all. Works like a charm to me
 
 
5 hours later…
11:23 AM
I agree - it is a duplicate. Worse - I was the one who asked it back then!! My apologies — Imray 1 min ago
^^^ wow. Just... wow
 
11:38 AM
comments today are "stellar"...
@gnat i don't have rights to post there. this is the next place where i can post programming related questions — Karthik Surianarayanan 6 mins ago
 
 
2 hours later…
user55340
1:58 PM
@JimmyHoffa Flags are often a "meh" and not touch it... unless I really agree with it. That said, being able to poke at the "what has 4 close votes and need another one" might be useful.
 
user55340
(you don't even need to hit /review for that... well, kind of... you need to go past it to get to programmers.stackexchange.com/… )
 
user20683
2:09 PM
Why do people insist on dousing themselves in so much Pachouli that they reak from 30+ feet away?
 
user20683
It's like a coked out raccoon going at my throat.
 
@MichaelT As for the last chat on contemplating about a new language to pick up... I decided to go for Java again. I think known Devil is better than unknown God... Focusing more on J2EE and I find it quite interesting to do it all over again :D thanks for the support!
 
user55340
@bonCodigo Java isn't bad. The other thing is that if you understand the library well enough, grabbing another JVM language can be fun too for when you want to dabble - you don't have to learn another library structure.
 
user55340
As awkward and ceremonial Java can be, that is often a good thing for team work (Ruby scares me working with a team every time I think about it)
 
user55340
@JimmyHoffa In the mean time - enjoy: programmers.stackexchange.com/questions/136210/…
 
2:16 PM
Either I am going to make with it or break with it. But I am just going to grab it and stick with it. I feel quite energetic after taking few refresher courses ;) Till next time Ciao. Thanks to @GlenH7 too
 
user55340
Colorblind co-worker comes in "Happy St. Patricks day!" (he's wearing brown and tan... which could be considered irish... well, black and tan)
 
user20683
@MichaelT Irish Coffee, very programmer.
 
user41796
@WorldEngineer 'cause the other aroma they are masking it likely to get them in trouble.
 
user55340
@GlenH7 Depends on the social group the person belongs to.
 
user41796
@MichaelT Hmmm, I hadn't heard of that but that would possibly explain one of the co-workers who uses that scent quite a bit.
 
user55340
2:30 PM
@GlenH7 There are only so many social queues that person can give visually. Switching to a scent based one allows for another range of expression that can help with the initial determination of "is this person into that" without being visually blatant about it (which can cause other awkward situations).... ahh, the joys of living near SF.
 
user20683
@GlenH7 Violently allergic to that other smell. Also why I don't like beers with loads of hops.
 
user55340
Btw, I blame @Ampt - I saw a snow plow salting the roads this morning. So I blame him for all the bad weather I'm about to get.
 
user20683
@MichaelT DC's gotten like 14+ inches of snow
 
user41796
@MichaelT And I don't think there is anything exclusive between what I think is the primary crowd for that scent and the alternatives. :-)
 
user41796
@WorldEngineer I honestly have never heard of anyone allergic to the scent, and I didn't realize it was possible. Thanks for expanding my horizons for the day. :-) Sorry to hear of the allergic response though, that could be challenging at points on a college campus.
 
user20683
2:35 PM
@GlenH7 The smell makes me wretch and feel like my organs want to be outside.
 
user20683
I'm generally sensitive to smoke
 
user20683
resin heavy smokes like sage for instance, are the worst.
 
user41796
@WorldEngineer ironic in that it's proven very effective in counteracting nausea for cancer patients.
 
user20683
@GlenH7 I suspect it's more the smoke than the drug but I could be wrong. I steer well clear.
 
user41796
This is the first article on the allergy I have found that appears to be researched. internalmedicinenews.com/single-view/…
 
user20683
2:39 PM
@GlenH7 Won't load for me at work, will read later
 
user41796
Might be keyword blocked. Summary is that the allergist ran an ad-hoc skin prick allergen test on a sample of patients who had reported sensitivity. Vast majority of those reporting sensitivity developed mild to severe wheals in response to the skin test.
 
user41796
And the rest of the article went behind a registration wall. Oh well. If I were to abuse the statistics, since 15 of 17 patients had an allergic response that means nearly 90% of the population is allergic. :-D
 
user41796
 
2:56 PM
TIL that cannabinoids in cannabis are effective at killing MRSA, a drug-resistant bacteria. Thanks, guys.
 
user41796
@ThomasOwens always some sort of education going on around here...
 
user55340
@ThomasOwens Your site blocks imgur, but not cannabis?
 
@MichaelT Yes.
Actually, I use HTTPS on Wikipedia. It has problems with HTTPS.
Although they've managed to figure out how to block HTTPS GMail.
 
user55340
@ThomasOwens Yep. With https, the virtual site you are connecting to is part of the encrypted https link... and its harder to peak into the data.
 
user41796
@ThomasOwens The other services like G+ aren't always blocked. Sometimes I would find chat available in one tab but closed in another.
 
user55340
3:00 PM
Not impossible, but it involves setting up a man-in-the-middle and then starts getting into the "oops! We looked at something that should have been encrypted" problems.
 
@MichaelT I don't know how they managed to block GMail. It doesn't get to the block page from WebSense. It's refused by the proxy.
 
user55340
Gmail has a distinct ip address despite being https. They can block the IP.
 
Specifically, Chrome says "ERR_TUNNEL_CONNECTION_FAILED".
Ah. I thought all Google services used a block of IP addresses.
 
user55340
(one of the products at Netapp back when I worked there was Netcache - a caching appliance... and I learned all about https, filters, and ways around them)
 
And you couldn't tell by IP address what Google service was being used. Just that it was something Google.
 
user20683
3:01 PM
Whiteboard sometimes loads, sometimes not.
 
user20683
I suspect age of browser
 
user55340
Name:	googlemail.l.google.com
Address: 74.125.225.86
Name:	googlemail.l.google.com
Address: 74.125.225.85
> www.google.com
Server:		172.25.16.20
Address:	172.25.16.20#53

Non-authoritative answer:
Name:	www.google.com
Address: 74.125.225.83
Name:	www.google.com
Address: 74.125.225.84
Name:	www.google.com
Address: 74.125.225.81
Name:	www.google.com
Address: 74.125.225.82
Name:	www.google.com
Address: 74.125.225.80
 
user55340
If they block 86 and 85, they've blocked gmail but not google.
 
user41796
GMail sign-in signs you in across all of the services. I used to have problems signing out because it was trying to sign me out of YouTube which had been blocked by the company.
 
I get error messages sometimes when logging in to site at work with my Google account because of that. The sign in usually works, though.
 
3:09 PM
@MichaelT now that's what I wanted 10k for. Great larf to start the morning with
 
user41796
@ThomasOwens - apropos of nothing, if you're into dubstep, Scrillex's new album Recess is pretty good. I was listening to it last night on iTunes radio since they were doing a free preview
 
@GlenH7 Skrillex is in New Hampshire. The day I leave for Seattle. He's with...DJ Snake, too.
 
user55340
@JimmyHoffa Give me a bit and I'll hunt up the MSO question with all the "great deleted posts of the past" that has some P.SE links too.
 
user41796
@ThomasOwens gah!
 
But it's OK. Avicii on April 10, Hardwell on April 16, and Krewella on April 29.
 
user55340
3:12 PM
123
Q: Popular Deleted SO/SU/Programmers Questions list

systempuntooutStack Overflow What was your first home computer? What's your favorite "programmer" cartoon? Programmers' last word How can you tell if a person is a Programmer Worst UI You've Ever Used How do you clear your mind after a day of programming How do you deal with a program that's bec...

 
user41796
That's like the time I flew into Vegas, staying at the MGM, and Tiesto had played the night before.
 
user41796
Hardwell was playing the night I landed but the tix were $100 / head
 
user55340
Not too many P.SE ones there... more fun to dig through old meta answers.
 
These were all <$30 tickets.
Actually, Avicii might have been $50 for floor. And floor > all.
 
user41796
I like Hardwell and enjoy when he covers the podcast for Tiesto. But I couldn't drop $100 for a concert at that point.
 
user41796
3:13 PM
Vegas and MGM probably accounted for $75 of that ticket price....
 
Yeah. Boston's probably generally cheaper than Vegas, even at House of Blues.
 
user41796
@MichaelT Now I have a reason to get to 10k on SO
 
Plus, they are weeknight shows.
Work the next morning is always a fun time.
 
user41796
Everything in Vegas seems ridiculously overpriced. More so than it used to be. You used to be able to pick up drinks & entertainment somewhat cheaply, but not anymore.
 
@MichaelT I've got Announcer for "Where to start": you managed to make an appealing title for askers to click
recommended reading: Where to startgnat 2 days ago
 
user55340
3:18 PM
@gnat First one in a year.
 
user55340
(and IMHO, the only one of them worth reading - meta.programmers.stackexchange.com/help/badges/54/announcer ... well, the contest wasn't bad, but temporal)
 
user55340
The next challenge is to get the only M.P.SE booster (300 unique IPs)
 
@MichaelT with Where to start Booster is going to be easy I think :)
 
I just turned my first screen pink
 
user55340
@JimmyHoffa doesn't it feel good?
 
3:23 PM
@MichaelT Clearly the snow plows bring the snow. Otherwise they wouldn't call them snow plows, am I right?
 
user55340
@Ampt Yep. And I blame you.
 
@JimmyHoffa pink? You might want to get that checked out.
 
user41796
@Ampt The consultancy has already affected your logical abilities
 
user41796
@Ampt Don't be jealous. You only wish you knew.
 
user55340
3:26 PM
 
@GlenH7 smells likes.... responsibility
 
user55340
@JimmyHoffa here's another pink one for you to laugh at - programmers.stackexchange.com/questions/46835/…
 
user41796
@Ampt True. Try to avoid it for as long as possible in your life.
 
user55340
@GlenH7 The magic number is 612
 
3:48 PM
2 days ago, by Jimmy Hoffa
I can't do it. I'm going with design option #1 and #2 -> I'll make a retriever instead of a repo, it will allow me to avoid breaking the Law of Demeter on reads of the model. The update code will still live just in the messages. Pondering implementing a deep clone on each entity in the model so the reads can call Clone() and always return new instances so no two readers have access to the same object and none can access the internal model instance directly other than the repo and update messages
2 days ago, by Jimmy Hoffa
the "retriever" will just be a read-only repo
@MichaelT @GlenH7 seem reasonable?
Or do you think the law of demeter is ignorable in this case?
 
I am wondering, does ignoring the Law of Demeter give you tangible benefits in this case?
 
@RobertHarvey Removes the need to write and maintain a repo
 
What are the possible perils of ignoring it?
 
user55340
The law is a guideline of an ideal. Sometimes the ideal gets in the way and should be broken. Yes, it introduces deeper coupling, but if you're going to be coupling it that way anyways, its not worse.
 
but does so at the costs of what the Law of Demeter warns: The object graph you're traversing into may be changed and all consuming pieces will then have to be changed accordingly
 
user55340
3:53 PM
It means a change in one class may break another... though not be immediately obvious.
 
Whereas a repo puts the object graphs structure knowledge all into a repo. At the cost of a repo's worth of maintenance over the lifetime of a product; never an insignificant amount of maintenance
 
So its significant maintenance either way.
 
@RobertHarvey Aye
 
Which way is more convenient, maintenance wise?
Design wise?
 
@RobertHarvey Repo
 
3:55 PM
There you go, then.
 
The repo seems right to me... I've minimized it by making it only house read operations
All update operations logic will be owned by the update messages that get passed around
 
Seems reasonable to me.
So, to be clear, the repo preserves LOD?
 
user55340
4:07 PM
Oh, I found I had this headset sitting around that I've brought into work:
 
@RobertHarvey yes; it ought anyhow.
It also gives me the ability to ensure I return clones so consumers are always holding copies of the source model objects and never directly referencing the source model objects
Not certain how much value that will present since they're read-only anyway, but it's an option I could do with the repo, but not with direct model traversal
 
I don't presume to fully understand your application, but I do like the encapsulation that would be provided by a repo.
 
user55340
4:26 PM
That math migration problem is unsolveable:
 
user55340
0
A: Calculating the Location of a Point Relative to a Rectangle

Thomas J RiveraYou cannot without knowing the location of the rectangle. Namely, for any circle, you could always compute two points $(x,y)$ and $(x',y')$ in the circle, and two rectangles $R$ and $R'$ out side of the circle such that the distance between $(x,y)$ and each corner $r_{1},...,r_{4}$ of $R$ is equa...

 
user55340
(being unsolveable is an answer... and one that would have been a mess on P.SE - thanks for the migration @WorldEngineer )
 
Why is that unsolvable? Intuitively, it seems like it ought to be solvable. It might involve pi, but still...
 
user55340
There are multiple answers - there is no unique rectangle.
 
user55340
The trivial one is you get a rectangle at (0,1) and one at (0,-1)
 
user55340
4:29 PM
 
So you would also need to know the angles of the blue lines.
 
user55340
Just to clarify, nothing else is known, including the size of the rectangle. — Builder_K 2 hours ago
 
Wait, no. The angles of the rectangle are all 90 degrees. There should only be one rectangle bound by those three blue lines.
 
user55340
Flip it to any of the mirrors.
 
OK. So that makes... four?
What if you limited it to one quadrant?
 
user55340
4:31 PM
Yep. Now, you might be able to constrain it to a given quadrant with that... but there might be other math at work too.
 
user55340
You could also rotate the rectangle 90 degrees and have a valid answer too.
 
So that makes eight.
 
user55340
still... math problem, not programming problem.
 
Seems like you would need limits to solve it. A brute force algorithm, maybe.
 
user55340
Its probably solvable with the appropriate math and the right set of equations. But math is my weakest part of my CS background.
 
4:34 PM
@RobertHarvey You could write it in a dependantly typed language and don't need to brute force it
money says it could be encoded in the type system
 
user55340
(I'm still shaking my head over the idea that you can make a sentence to a single number)
 
(I have no idea what problem you guys are talking about, but if it's mathematically based and brute forceable, it's probably encodable in a dependently typing system)
 
Types in Haskell are that powerful? Or you're just pulling my chain?
 
@RobertHarvey Haskell reaches towards that but no, it is not dependently typed, there are some extensions and offshoots, but dependently typed languages are generally epigram, agda, coq, hol
any of those could likely do it
idris
ATS, Mercury (a prolog like dependently typed language - also a good approach for bruteforce solvable problems using constraints)
@RobertHarvey depends on weather or not your sir name is lucky
 
Did you see how much of my time that guy wasted?
 
4:39 PM
@RobertHarvey I saw bits but mostly ignored it; me thinks you were trolled
 
user55340
Well, good thing you're not a P.SE mod. You'd start trying to learn some haskell and then delete half the database.
 
user55340
speaking of which... go use some vtd @JimmyHoffa ;-)
 
@JimmyHoffa Yeah. That's not hard to do. I'm so gullible; I assume good faith unless someone proves otherwise. I suspected about halfway through the conversation I was being trolled.
Good thing I have strong opinions. My response is simply, nope.
 
user55340
Btw, congrats on the tag badge: programmers.stackexchange.com/help/badges/76/…
 
Ah no congrats for the awesome badge I just picked up programmers.stackexchange.com/help/badges/14/…
Fitting badge considering I try to be a full stack dev
 
user55340
4:44 PM
@JimmyHoffa I didn't see that one... congrats to you too! That's a non-trivial one on P.SE
 
Yeah, I think that's the first badge I've earned that I actually appreciated; much harder than any other badge I've got I think
 
5:07 PM
Your logic is reasonably sound, although your numbers are off by at least three orders of magnitude. I would expect a traditional RDBMS to readily handle millions of records. — Robert Harvey 29 secs ago
These kids nowadays. They think rice rockets are muscle cars.
 
@RobertHarvey Millions? I've managed billion row tables fine in MSSQL, you just have to design the tables right, and they'll churn reads and writes with perfectly good response times
 
@JimmyHoffa He was saying thousands, and suggesting NoSQL as a fix.
 
Millions? I've managed billion row tables fine in MSSQL, you just have to design the tables right, and they'll churn reads and writes with perfectly good response times even in high load public facing production systems. I have personally seen this work fine in more than one job. So millions? Doesn't even scratch the surface of what an RDBMS can do. NoSQL has it's place and it's purpose, but just because you have lots of records doesn't mean you need NoSQL. — Jimmy Hoffa 13 secs ago
 
Works for me.
The NoSQL databases like BigTable are supposed to be more scalable.
 
apparently my school is having a xeon phi supercomputer donated. Am I the only one who thinks that it may be too little too late?
why would we want to use a supercomputer when EC2 instances are cheap and plentiful
 
5:14 PM
Apples and oranges.
 
Working on 6 terrabyte MSSQL databases has taught me a lot about what RDBMS can and can't do; many people give RDBMS too little credit and many others give it too much. Problem is most people are working off hearsay; not a lot of people have genuinely seen a well-designed RDBMS system reach it's capacity. Many have seen crap designed ones fall down but that has nothing to do with the RDBMS capabilities, rather only speaks to the implementors
 
NoSQL is an excuse for young 'uns to not have to learn SQL.
 
@RobertHarvey often times
but that's not the whole of the issue
 
Why would I learn SQL when I can do everything in a giant key/value store?
and write loops all day long.
 
your statement is an excuse for many old-hats to stick with the date that brought them regardless of where NoSQL may make sense
 
5:16 PM
Idk about you guys but I just store everything a text file in my dropbox folder
backed up and handy!
database shmatabase
 
@Ampt obviously, where do you think the 6tb database files MSSQL works with are held? Dropbox, duh.
 
@JimmyHoffa well if you have 6tb I would suggest google drive
99.99 a month for 10 TB so you've got space to grow!
 
@Ampt @RobertHarvey Project for either or both of you: Answer that question with:

You've got relational data (it sounds like - frankly you told us of a relational schema, but didn't talk about the actual data's purpose and usecase). If you had non-relational data or non-transactional data then you should start thinking about NoSQL. But the bigger question would rapidly be, what *kind* of NoSQL would even make sense?

<line break>
Document Store
- bla bla bla about what it's for and how it might help him etc
It'd give you both cause to go look through and come up with a summary of each of the varied types of NoSQL
If either of you want to spend some time reading up and summarizing all the NoSQL types, here's your opportunity to get free repz for the learning
 
user41796
@Ampt - there is something to be said for having the supercomputer on site instead of hosted in the cloud. That kind of assumes you have admins and data center space that's already a sunk cost. There is a psychological shift when it's on premise and running anyway, so the administration can let all sorts of experiments run instead of doling out X $ of proc time out in EC2.
 
also a small summary of ACID and how/why it matters and why it's lost in NoSQL implementations (or which parts are lost in each type of NoSQL)
 
user41796
5:24 PM
Essentially, it's a joint asset vs. a per group funded asset.
 
My only point is that some folks reach for a key/value store first, but history favors the SQL-ist for most applications.
 
@GlenH7 Doubtful it'd be "running anyway" when you look at the electric cost to run any of those most times
@RobertHarvey Granted, still good to go read about all the different NoSQL types and see what they are and aren't good for
I haven't the time to go read up and summarize them in an answer today
 
user41796
@JimmyHoffa I think it would still be cheaper to scrounge up funds to cover the electricity cost than it would be to cover processing time.
 
there's so much literature to be read on the NoSQL stuff these days...
@GlenH7 Plausibly, doesn't change the fact that it wouldn't be "running anyway" - they would probably treat it like classical mainframes, they'd turn it on when they had the project in hand with signed budget affording the given project time on the machine
 
user41796
@JimmyHoffa But the electrons yearn to be free and used for wild, scientific pursuits! Don't deny the poor things just because they're negatively charged. They're constantly in search of something positive.
 
user55340
5:31 PM
There's also the "we're doing massively parallel OS design and need to try reloading the OS on the system" which doesn't work too well in the cloud.
 
@GlenH7 Now you're just sounding as crazy as me. I expect the next thing to come out of your mouth to be something about monads at this rate
@MichaelT Beowolf cluster? You had to go there?
 
user55340
Or "I've got these 5 different ways of approaching problem XYZ in massively parallel systems" what do each of these architectures do?
 
You've been beaten to the punch @MichaelT code.google.com/p/elasticwulf
 
user41796
@JimmyHoffa I don't think that allows the types of manipulations that I think MichaelT is suggesting.
 
@GlenH7 and a monad would? You so crazy Glen.
 
user41796
5:36 PM
I was listening to Yeezus the other day...
 
user55340
I recall UW Madison had ~5 different fundamental types of super computers (connection machine, clusters, something from IBM or Intel that had lots of lights as paths)
 
user41796
So they were doing real CS instead of just making up terms like monad?
 
user55340
@GlenH7 Yep. The thing is that condor (UW Madison - grew out of the team that had grown from the HTC projects) and others are just each one way to do HTC.
 
user55340
The Connection machine was just one way of doing it.
 
user55340
The Connection Machine was a series of supercomputers that grew out of Danny Hillis' research in the early 1980s at MIT on alternatives to the traditional von Neumann architecture of computation. The Connection Machine was originally intended for applications in artificial intelligence and symbolic processing, but later versions found greater success in the field of computational science. Basis Danny Hillis' original thesis paper on which the CM-1 Connection Machine was based is The Connection Machine (MIT Press Series in Artificial Intelligence) (ISBN 0-262-08157-1). The title is out of ...
 
user41796
5:44 PM
Ever have one of those afternoons where you just want to start saying cuss words over and over again? Mine just went that way.
 
@GlenH7 No, I do occasionally have those afternoons where I just start saying cuss words over and over again though.
 
user41796
@JimmyHoffa too many coworkers in close proximity
 
It goes over less well in the open office I work in now than when I was in a closed office with other seniors previously
 
@GlenH7 Was peyote involved?
 
I'm pretty sure ability to string a slew of verbal filth is one of the key skillsets of any senior engineer
 
user41796
5:48 PM
@RobertHarvey Not on my side. Possibly on the lenders side.
 
user41796
Refinancing my home mortgage
 
user41796
Looks like the initial estimates had left out a "needs to be paid soon" estimate for property taxes.
 
There's a lender called Yeezus? Sounds like a doctor named Sneezus.
 
@GlenH7 It's impossible to talk to any banker without some portion of costs being conveniently unannounced until the ink is dry
 
user41796
@JimmyHoffa No kidding.
 
5:50 PM
Just look at the APR disclosure. If the APR is significantly higher than the quoted rate, run like hell.
 
user41796
The worst part is I had talked with them on Thursday about these numbers.
 
Such as it is, the way a person shares information is terribly colored when they view others as commodities to be extracted as opposed to people
 
user41796
@JimmyHoffa it's through a local credit union, and I gotta admit the folk are nice. This isn't "just a transaction" from the way they have been acting. But this last minute change puts me in a helluva bind.
 
@GlenH7 gotcha. Yeah I do the same; working with local credit union for years on any financial transactions of import. Definitely a much more customer oriented folk.
holy shit I just cast a DV on a legal question
it felt so good
I'm going to do that again
hello ...
...and I'm out of DVs
...want more DVs....
 
user41796
6:07 PM
@JimmyHoffa welcome to the addiction
 
user55340
You poking at the 10k "this is almost closed" sniper page?
 
@MichaelT I did a bit
then I saw a licensing one in there and just decided to follow the tag to the honeypot
 
6:18 PM
14
A: Now that 20k users can delete answers, should we increase the number of daily delete votes?

Marc GravellTaking out the trash is indeed wholesome; this is being implemented such that you get an extra delete-vote (up to a maximum of 30 per day) for every 1000 rep after getting access to the moderation tools (typically 10k - I can't remember if that varies on beta/SE sites, etc) - so someone with 22k ...

^^^ gain repz for more delete votes
 
@gnat 10k was enough work; I'm not holding my breath for any more milestones
 
@JimmyHoffa believe or not but after 1k extra rep, you will likely feel that +1 del-vote motivates quite substantially
 
user41796
@JimmyHoffa 20k is awesome...
 
I'd feel guilty for not helping the site out more than I do with more review votes and answers written except that I've created plenty of quality content for the site I think; so if I write no more answers and do no more review votes, I figure I've been plenty net positive and deserving of the entertainment of looking at trashy deleted content heh.
 
user41796
immediately being able VTD on down-voted crap feels so good.
 
6:23 PM
@GlenH7 yeah... but having to take the time to write answers to get 20k... ugh... I'll stick to only writing answers when a question urks me enough to feel I have to say something, like this:
23
A: Why is an anemic domain model considered bad in C#/OOP, but very important in F#/FP?

Jimmy HoffaThe main reason FP aims for this and C# OOP does not is that in FP the focus is on referential transparency; that is, data goes into a function and data comes out, but the original data is not changed. In C# OOP there's a concept of delegation of responsibility where you delegate an object's man...

When I run across those questions I can't stop myself from answering. Other than that, I don't think I'll focus on rep-whoring anymore considering it's 5-10k from any more boon for the rep-whoring
I am however curious to see how the shortening of my rep effects readers. I think when they see "9930 rep" vs "10.1k rep" next to a user, there's a mental step they make to think more highly of one vs the other even though they're very close
 
user41796
@JimmyHoffa I am of the same opinion. My impression is that it took longer to get to 10k than it did to get to 20k
 
user55340
@GlenH7 Or cleaning out ranty answers.
 
user41796
I have more answers to pick up residual up votes, but I don't think that made a significant difference.
 
6:54 PM
I continue to be baffled about Gnat's conception of question duplication.
 
user41796
@RobertHarvey It .... can be a stretch.
 
user41796
I think he still prefers to VTC as duplicate instead of too broad | primarily opinion based | custom reason.
 
"What size should my shoes be?" Duplicate of: "Does this dress make me look fat?"
 
user41796
@RobertHarvey That one arguably may be answered in the duplicate since the logical answers are never correct. :-)
 
Yes, but common answers were never a criteria for closing as duplicate.
Even though some community members see "duplicate of" as a synonym for "it's answered over here."
That was never the case.
 
user41796
6:59 PM
@RobertHarvey I think he sees it that way, and I'll vote that way as well if it's clear how the answer addresses the new question.
 
The questions need to be damn close to each other for that to work.
 
8
A: Does the new guidance on duplicate questions suggest closing a question as duplicate, if the original answers the OP's question?

voretaq7My two cents: Is it OK to close a question as a duplicate, even if the questions aren't the same? Absolutely. We do it ALL THE TIME on Server Fault when we send people to the Canonical "What I've been hacked! What do I do now?!" question, or the various Canonical Capacity Planning quesitons. Th...

 
@gnat He's talking about canonical questions.
Canonical questions are still a bit of a half-baked concept. Unless they are structured very carefully, they tend to get closed on Stack Overflow.
 
Abby T. Miller on March 17, 2014

Welcome to the Stack Exchange Podcast #56 recorded on Thursday, March 6th 2014, aka the 4th of Adar II 5774, aka the second day of Lent. Today’s podcast is sponsored by Patent Trolls of America. Today’s guest is Micah Siegel, Senior Patent Advisor at Stack Exchange and Professor Emeritus at Stanford.

But first, Community Milestones!

We’ve already talked at length about The Workplace, but it should be noted that the Workplace community has just graduated. They are now a fully-fledged site, so go check out their design! …

 
I'm glad ServerFault managed to figure this out. But if you look at the questions that Vortaq linked, you'll see that the process is pretty elaborate for making an original question a canonical one, and we've never done it well on Programmers, AFAICR
 
7:08 PM
@GlenH7 Our IT managed to lose every single email in every single persons drive over fall break
I give the supercomputer ~3 months of actually running
then ~25% uptime
 
user41796
7:20 PM
@Ampt That's 25% more time than they would have had before. #glasshalffull
 
user41796
@RobertHarvey We have kicked the idea around a bit, but it's never quite stuck. We have the "what every web dev should know" question and the "I've inherited 200k of spaghetti code", but the other proposed ones lack a knowledgeable curator.
 
So I'd like to pose a poll-like question for developers to find out if my work development environment infrastructure is common or not in the greater corporate world. Is there an appropriate SE place for such a thing? SO? Meta SO? Programmers? Workplace? Or one of the chats?
Appropriate place to pose such a question I mean....
 
user41796
@PeterTirrell Chat. Generally not appropriate for main or meta sites.
 
user41796
Polls default to close as "too broad"
 
@GlenH7 Definitely not appropriate for meta sites. No main site accepts polls, either.
 
user41796
7:25 PM
That said, each chat room has different groups presents. So if you're looking for a particular type of dev, that may affect which chat room(s) to ask in.
 
@PeterTirrell The real question is why does it matter? Do you think that your stack is missing something/inefficient, or are you just curious?
 
@RobertHarvey The anwer to both being "no" ?
 
user55340
@RobertHarvey immutability is core to several OO language bits (such as Java String, BigInteger, wrapper classes...)
 
user55340
If Integer was mutable, BAD things happen.
 
Learning VIM
 
7:33 PM
Just curious :-)
 
@RobertHarvey I think software engineering simply can't have "canonical"
 
user41796
@Ampt emacs FTW
 
user41796
@JimmyHoffa quit trolling Thomas and I.
 
No good way to search for files outside of Cscope, and if I want to view a file within cscope it opens vim so...
 
user41796
:q! is your friend
 
user41796
7:34 PM
forces a quit without saving
 
user55340
I work at a hard code vim shop. Even the sublimer is learning vim. Two of my co-developers are hardcore editor geeks.
 
We're entirely on a virtual development infrastructure, i.e., we log into VMs from thick desktop clients and do 100% of our development work on the VMs. Normal .NET stack. But the experience isn't great and I feel like the VMs are underpowered. Just wondering if that's the norm elsewhere or if I'm special :-)
 
@GlenH7 I mean it. It's the pace of our space. SO could have canonical because there are canonical ways to write bits of code, sure. But of conceptual things we work through as engineers? The canon is always moving for us. What's canonical one year isn't the next in terms of design. 10 years ago RDBMS would be the canonical back-end system for a CRUD site; it just isn't so anymore.
 
user55340
You're special. I've rarely seen that... the only time I saw that it was remotely practical was where it was an occasional iOS development and people remoted into a single mac (rather than buying another machine for each dev).
 
We deal with issues that are never black and white, that's what P.SE is for
@MichaelT Not that hardcore if they still haven't realized they should be using EMACS...
 
user41796
7:37 PM
@JimmyHoffa I'd argue; but my DB instance just pitched a fit. Gonna be It's one of those days.
 
Wow... that was... surprisingly efficient
 
@GlenH7 Just imagine a glass of famous grouse
 
user41796
@PeterTirrell I'm with MichaelT on this one. That's a bit unusual. But it shouldn't be that big of a deal to allocate more resources to the VMs assuming the main server isn't over-allocated.
 
user41796
@JimmyHoffa Gonna need it. Have some at home, actually.
 
@PeterTirrell ask for more power!
 
7:38 PM
@GlenH7 which they always are. Thank business rags for pointing out that any machine can become 2 or 4 or infinite machines when you install VM on them
 
user55340
@JimmyHoffa I told 'em to get back to ed
 
remind them how much every minute of waiting costs them in developer time ;)
 
@Ampt @GlenH7 @MichaelT thanks! Hah, oh, I'll be definitely seeing about getting more resources, but I wouldn't bet the main server isn't over allocated
 
people just never bother to imagine they're using real resources with virtual machines. I mean, if the machine is virtual, clearly the resources are too, so you don't have to worry about exhausting them...
 
user55340
@GlenH7 that said, remoting into a VM and pulling over the windowing envirment sucks network resources big time. Trying to do a debugger to a remote VM... not fun.
 
7:40 PM
if you wait for 5 minutes every hour for the VM, and you make 60k per year, thats ~5k a year they are paying you to just sit there and wait
 
Good to know...I think I was maybe hoping that we were on some sort of bandwagon, even if it wasn't great from my experience. Not sure if I feel better or not now!
 
user55340
I've seen people try to do it (citrix style) back in the day with remote / not completely trusted outsource shops. Got expensive for network and a sizeable enough system to actually work on.
 
(double check the numbers if you're gonna quote me, that's just a back of the napkin calculation)
 
Hi
is there any moderator here?
 
@MichaelT thats why 10g switches were invented
 
user55340
7:40 PM
On the other hand, I have seen people in a Windows shop be forced to run linux on the local machine as a VM to get the proper enviroment for production minicing
 
user41796
@Revious depending upon the site you need, yes
 
user55340
@Ampt Santa Clara to Hydrabad != fast link
 
@GlenH7 Programmer
 
@MichaelT ouch. no. I'm talking VM into a server rack inside your building
our SVN server is about 30 feet from me in the server closet and I get stupidly fast speeds when I'm pushing/pulling
 
user41796
@Revious We have one in the room presently, can ping others easily, and you have multiple 10k users in here. What's up?
 
7:41 PM
@MichaelT it probably works better depending on the development stack. MS stack? Not so great. Anything where the software runs on nix and you can do the dev work in VIM or EMACS? Probably works just fine having nix VMs people just putty to
 
@JimmyHoffa just do HDMI over cat6!
 
or using an X client
@Ampt ...virtual machine remember, do you plug the physical cat6 into the virtual nic?
virtual HDMI: Your imagination at work.
 
found that the other day and it was a pretty cool read
 
@Ampt VMWare trash, I'm not reading that :P
Does it show how to use multitenant physical display over a single NIC? :P
 
@JimmyHoffa aww is someone stuck in the .net stack?
it's ok little guy, we'll help get you out
 
user55340
7:44 PM
@JimmyHoffa text works nicely, untill you need to attach a debugger to a java process running in an app server.
 
@Ampt That troll would be more effective if I disliked VMWare for any reason other than it's pointless, painful, and harming the industry as a whole; not at all to do with working in MS
@MichaelT Like I said, an X client ought work well enough; those are fairly efficient on network, no?
 
user55340
Its hard enough to get good programmers when doing such an approach for cost cutting outsourcing. Much less one that can work with the internals of jdb from the command line.
 
@JimmyHoffa how is vmware harming the industry as a whole. I've never had any problems
well.. I take that back, but that was more a bad database connector setup than VMWares fault
 
user55340
@JimmyHoffa Until you hit a bitmap from an app that isn't rendering just text and vector graphics... like a web browser.
 
user55340
(and then it sucks golf balls through a narrow straw)
 
7:47 PM
@MichaelT makes sense
 
Thank you again and good answer. I have to digest it a bit. I also think that polymorphism and (not inheritance, but) derivation also figure quite centrally in functional languages. — Hyperboreus 7 mins ago
Polymorphism in FP?
Perhaps he's thinking of dynamic typing. What is "derivation?"
 
how is VMWare bad? Anyone? This is the first I've heard of VMWare being anything but positive
 
Nah... he's confusing things
You can't read anything on FP without running into terms like "parametric polymorphism" "higher kinded polymorphism" "rank-N polymorphism"
 
user41796
@Ampt I like it. Kinda pricey for enterprise level stuff. But it's downright amazing in how it can migrate things around.
 
but those are so far off from subtypal or prototypal polymorphism and he likely doesn't know the FP languages that would have caused him to learn what those types of polymorphism are
so he's just grasping at straws as "polymorphism is important to FP" but not really knowing what that means or how it is
 
7:50 PM
I'm not sure I'm fully aware either. But I do know that the kind of polymorphism that we have in C# has little or nothing to do with FP. Generics, maybe.
 
@Ampt It's uninteroperable and puts complex engineering decisions in the hands of IT groups in many places
@RobertHarvey Exactly; you know they have nothing to do with eachother, I suspect he doesn't
all the VMWare automation that exists is pure garbage, very old APIs that are clunky as shit
 
user55340
@JimmyHoffa Note that my former Cali employer was Netapp. Netapp is in competition with EMC. EMC bought VMware in 2004 and so Netapp wouldn't touch it (paying a competitor for products, or needing to get the professional services from the competition to come in and tinker with VMs on your internal network... not gonna happen)
 
01:00 - 20:0020:00 - 23:00

« first day (1291 days earlier)      last day (3693 days later) »