The Whiteboard - 2015-02-26 (page 1 of 2)

« first day (1637 days earlier) ← previous day next day → last day (3356 days later) »

user55340

12:42 AM

@WorldEngineer At Employer^^, QA would take the test cases that the devs wrote, and follow them. No 'monkey' testing. No 'user' testing. Follow exactly the test case the devs wrote. The problem there is that devs tend to be bad at writing test cases that fall outside of the expected path. So our test cases would pass (we test them ourselves) and QA was just a rubber stamp of "yes, the things the devs said would work work."

1:45 AM

evening all

user20683

Evening

1:52 PM

@MichaelT Don't really need to show up to work to do that job

@MichaelT QA could then be replaced by a automated test rig

user55340

2:10 PM

@ratchetfreak QA's sole responsibility was working on automated acceptance tests as a long term project... With no end in sight.

user55340

They weren't developers and were trying to write Cucumber.

@MichaelT How do you write Cucumber without knowing Ruby?

user55340

Anyone who knew Java was a developer. They knew excel and vb... Kind of.

user55340

The "I took a semester of that back in college " knowing.

user55340

So, they showed up, clicked buttons showing that the devs did what they said. When there is a bug that got to production, they say "the devs didn't give us a test for that." When there was downtime, they tinker with excel and vb in a doomed application.

user55340

2:18 PM

And when I get to a real keyboard, I'll tell you of the unit testing goal that we laughed at.

2:32 PM

100% code coverage? I love that goal.

it means I can promptly write off the requester as worthless.

user55340

Unit testing on a version of progress that didn't support it was also funny.

user55340

But there was a reporting requirement too. Ping me at 6pm cst and I'll rant about it for a bit.

user55340

3:09 PM

@GlenH7 tossed a layman question for people on engr.se. Please feel free to modify or whatnot.

user41796

@MichaelT Will be interesting to see how that one goes

user41796

There's a contingent that wants to see "softer" questions (education advice, etc...) and I think your question is a pretty high quality one.

user55340

The bridge in question was in the south, 100+ years old, with rotted beams that you could stick a rod through. And was possibly carring oil tank cars.

user55340

I was looking at it from a "what do non engrs want to know and are likely to ask"

user41796

@MichaelT This user appears to be following the same tack. Some of his questions have been misses, but others have been pretty good. My understanding is he's trying to walk that thin line of constructive click-bait type questions, which is understandably difficult.

user41796

3:18 PM

This one of mine picked up a minor shit-storm in the commentary. People got too wound up about the original title and wording so I had to change quite a bit of it. It's now +7/-3 but at one point it was +3/-3.

@GlenH7 I wish funnier titles like "ouch! what hit me?" weren't so frowned upon

having a catchy title does not mean it's a dumb clickbait question

user41796

Agreed.

user41796

But there was one particular user who got really bent out of shape by it. Although there was another who immediately said "lighten up" about it

@MichaelT I'm not sure about that one, It sounds like a "who is the guy in the government that I should talk to?"

and is going to depend on jurisdiction

user41796

@ratchetfreak Understanding what agency(ies) have authority in regulating bridges could be of value

3:23 PM

@GlenH7 I think the best title is actually: "Ouch? What hit me? Quantifying the impact of a martial arts strike"

user41796

@ratchetfreak I've argued the jurisdiction aspect on other questions and current community consensus is that they can be tolerated so long as the asker provides a jurisdiction.

I agree the second part is too vague.

Which @MichaelT didn't...

user41796

I personally don't agree and I think it will become problematic. But I expressed my vote, and I'm happy to go along with the community to see how it plays out

user41796

@ratchetfreak FTFY. :-)

user41796

3:27 PM

@durron597 I liked it too. Especially because of the visual it provided. I had hoped it would link into a Q&A about the various quantities and their units that are involved in a strike. HDE's answer came closest to what I was hoping for.

user55340

It's one of the "see where the edge of scope is over here"

... and get shamelessly upvoted by the p.se chat? :P

user41796

@enderland meh, the site needs more activity and votes anyway. Little harm in my book. :-) Obviously, don't get carried away and trigger the serial up voting reversal algorithm.

user55340

Hard to do with one question. But hot wouldn't be bad/awful with a good question that properly shows the scope of the site.

user41796

Yep

user41796

3:41 PM

There's a couple of active civils on the site that I would like to see put an answer in place. Preferably without the condescending tone of the current answer

Ohhh... Drama.

Anyone in here familiar with BDD?

user41796

@RobertHarvey Probably about at the same level as you are

I'm genuinely curious about this question:

1

Q: How to use unit tests when using BDD?

I am trying to understand BDD. I've read some articles and as I understande BDD is "the next step" from TDD. I say that because I find both to be very similar, and as I could read in this artcile, BDD was born as an improvement from TDD. Great, I really like the idea. There is one practical poin...

unit-testing bdd

Presumably BDD works the same way as TDD, in the sense that you write the tests first.

I think this guy might be laboring under the invalid assumption that you can "grow" a design from red, green refactor.

user41796

@RobertHarvey I think the problem here is the assumed technical linkage between TDD and BDD

user41796

The only "next step" I can see is from the perspective of "how do we write better software that meets the need of the business?"

3:49 PM

I don't claim to know the first thing about BDD, except the phrase "it's TDD done right." "Instead of writing tests you should think of specifying behavior."

user41796

BDD to me is more about mapping software constructs to business objects

> When your development is Behavior-driven, you always start with the piece of functionality that’s most important to your user. I consider this phase as taking the developer hat off and putting the user hat on. Once you’ve specified the user needs, you put the developer hat back on and implement your specification.

user41796

and you can more naturally define what the correct behavior is

Well, I've seen BDD frameworks that allow you to write "code" like this:

describe User do
  it "lets me assign a name" do
    user = User.new "Paul"
    user.name.should == "Paul"
  end
end

reading wikipedia, it seems that they expect business people to write the names of the unit tests for you, which is a pipedream.

user41796

3:52 PM

I haven't implemented a "one true BDD" system, so I'm not familiar enough with it to comment. But I'm notorious for stealing the gist of good ideas and using the good parts within my own work.

Actually, it's interesting how DDD and BDD are solutions to the problem that TDD doesn't naturally "grow" a design.

BDD focuses on use cases. DDD focuses on domain objects.

BDD lets you write your use cases in a way that makes more "use case" sense than in TDD where you might write them in a technical sense

so you can write them like

When a user runs cmd
then outcome 1 happens
and 2 cookies are built

Sounds like a user story, but in greater technical detail.

Yeah

user41796

I can haz cookie?

4:03 PM

I was super skeptical initially of this because it's hard to really "get" initially

Because of the "Test-First" rule? Or something else?

it adds several layers to the code

you have your "when a user runs cmd" clauses which then all have definitions programmatically

since you have to define what code to run for that line of your test

But it makes writing tests a lot better since you don't have to worry about translating your "what is going to happen for real" into code, since you write the code for that once and then can write english tests

Ok, so the frameworks for this are really important, since they make the use of that fluent interface style that becomes possible in the code above.

Yeah, I'd assume so

we're using lettuce.it/tutorial/simple.html

if you skim over those code examples it's pretty straightforward

Hi guys, nice to see that you are talking about the question I wrote, maybe I can clarify it here

4:12 PM

@Oscar Did you read up? --^^^

I'm doing that now

@enderland That seems exactly like the Mockito language... "when / then / verify"

@durron597 well only so many words would make sense to use like that ;)

@enderland But it's the same concepts too

as a [some type of] user / i want to [do something] / so that [some result occurs]

4:16 PM

bah, natural language programming is a trap - even for unit tests.

So, this is what is confusing me. I found in many threads in stackexchange answers similar to this one: http://programmers.stackexchange.com/a/164758/7764 or this one http://programmers.stackexchange.com/a/245746/7764
In which they state that "BDD is not a technology, it is a technique" and that includes unit tests.

user55340

Btw, welcome @Oscar

thanks @MichaelT

@Oscar DDD is a technique, not a technology also.

The misunderstanding causes questions like "Where do I put my aggregate root," which kinda misses the point.

Great, we currently have BDDs as integration tests in our system, in the sense that our BA writes the statements and we implement it.
I do not feel safe because we have no unit tests, and I would like to have them. My question is: how do I create them? Should I write statements similar to the ones the BA write so the behavior of the method itself is described?

Or should I do "plain" unit tests, without any kind of behavior description approach?

4:20 PM

It seems to me, as you get closer to single methods and objects, your test descriptions will be more technical and programmer-oriented, and look less like use cases.

@Oscar My $.02: Remember the techniques exist to serve you and not the other way around

(In Soviet Russia, Techniques SERVE YOU!!)

Does that make sense? Sure, you will write the BDD tests to represent the use cases, but you're still going to have to break that down into methods and objects that are more specific, just like you would if you were doing ordinary use cases and software requirements on paper.

The only difference that I can see here is that the use cases are actually coded into the test suite (which is a good thing).

@durron597 true, I agree, I just want to avoid errors that are well known by the community :)

As I see it, the BDD tests are integration tests, written in fluent interface form.

hm, it makes sense @RobertHarvey

4:25 PM

Keeping in mind that neither I nor anyone else currently in here is an expert on BDD. :D

So take all of this with a huge grain of salt. If you want best practices, read a good book on BDD.

user55340

@RobertHarvey do I need to paste that youtube expert video again?

sure, I will :) But I'm sure this will be a nice first step

Thanks guys for all your help, the community is great :)

@MichaelT I feel a Rickroll coming on.

@RobertHarvey If you could post the reasoning you just had in the question, I would be happy to mark it as an answer

BDD doesn't mean you don't implement the tests

BDD means you abstract away the way you invoke the tests, to something more readable

but you still have to implement the tests

4:28 PM

Sure, @enderland. But as they are written by the business analyst, they are more integration tests than unit tests

@Oscar the implementations of the tests are written by the BA? or the actual test language itself?

No, the feature (test description) is written by the BA. The dev team will implement the tests

user55340

The Expert (Short Comedy Sketch)

2

Oh, yeah. That one.

Don't give me nightmares again.

If I can change the subject for a moment, has anyone else in here noticed a trend among employers that they now want you to know everything they need for their projects so that you can start running right out of the gate? In other words, they want you to have already learned what you need to know on someone else's dime. Is that a new thing, or has that always been the case?

Seems like there was a time when you could get a job just by demonstrating that you knew how to pick up the learning you need and how to solve problems and get the job done.

@RobertHarvey That's basically what Google said when I interviewed there several years ago. "We think you have talent but you're rusty. Go work for someone else to help you get back up to speed, then apply again"

4:37 PM

Egad, they were actually bluntly obvious about it?

@RobertHarvey They were explicit

user55340

Google can be picky. Others it's a pipedream for them to be that picky.

Career note: I was a Java developer from 2006 to 2008, I played poker professionally from 2008 to 2012, this interview was in early 2012

@RobertHarvey I wish all employers cared primarily about "are you smart and can you learn" ?

I think that may be more true for internal moves than external moves

@enderland That's why I keep offering you the chance to apply here

I'm a developer who pretty much has hiring power :)

4:39 PM

I read an article recently that said "Employers have gotten lazy about recruiting, and don't understand that the market has turned in favor of employees again, and that their current recruiting practices are no longer going to work, but they haven't figured that out yet."

I'm not sure. I don't think that employers have ever been trusting of employees to just figure things out.

but then again, employers are only slightly better than clients.

@durron597 yeah yeah yeah, want to move to the midwest??? :D

I told my boss outright in the interview for this job that I didn't know all the things I would need to know, but that I was smart and learned things quickly. I guess he believed me

@enderland Don't you mean "Midwest, US" (cc: @GlenH7 @JimmyHoffa)

user55340

@enderland want to work for the state of Wisconsin?

@RobertHarvey For what it's worth the only thing we've ever tried to find good people that actually resulted in promising interviews is face-to-face career fairs

careers.stackoverflow.com resulted in all crap, half a dozen recruiting firms have finally, after a YEAR resulted in the first on site interview, and I don't even like the guy that much.

I would take any regular in this room over every single person we've met through a professional recruitment firm in a heartbeat

4:45 PM

@durron597 I thought about it!

@durron597 I'm going to have to start doing those and networking and other similar things, because I think the whole sending out resumes thing is extremely haphazard and inefficient. For whatever reason, employers can't figure out what I'm capable of doing from my resume, even though they all agree that it looks impressive.

@enderland What about Texas over Iowa?

@durron597 no winter

@RobertHarvey How many times have you heard someone say "I do so much better in in-person interviews than my resume makes me look"

@JimmyHoffa I thought @Ampt is making a lot of money doing ~~nothing~~ consulting for a client that loves him right now

@durron597 I have heard that, yes. Not sure that describes me, unless I already know something about the company, how they work and what they want.

4:47 PM

@RobertHarvey are there local tech meetups down there? we have a state org that does tech meetups in a lot of cities

user55340

@durron597 there are crazy people who like winter. Some of them go shirtless to GB packer games.

I went to one and someone basically wanted to offer me a job (and afaik are trying to, just like a 3 month process at this point lol) and interview

I'm sure they those people would MUCH prefer someone with experience to a youngin like me anyways...

@enderland I will look into that. There are some things coming up; the local community college is holding a job fair, and all the aerospace companies go to that.

I have a bit of that resume vs capability problem, though until recently it was because of years of professional experience vs years of coding experience.

@RobertHarvey What I meant is that, that's why career fairs were good. I got to spend 5-15 minutes with a ton of people, I threw out 80% of the resumes we got but it still resulted in 6 interviews, which is more than we had in all of 2014

4:49 PM

Ah, I see.

One of which was the worst resume I've ever seen. But she was just bad at writing resumes.

@RobertHarvey I'd be shocked if there's not something like that in your area

@RobertHarvey I am in a similar position. Have you considered a portfolio of projects that you can show employers?

technologyiowa.org/en/events_services/techbrew is what we have (lots of them around the state)

@MichaelT It's 39 degrees today!

4:50 PM

@Oscar It's going to take awhile to put that together. Everything I've worked on in the past is company proprietary.

@RobertHarvey Have you considered doing the problems at Hackerrank? ;)

@Telastyn the problem is less of the employees and more of the whole process and more often than not non-technical managers which results in people being poorly evaluated

@durron597 Already doing that.

@RobertHarvey I know, that was what the smiley was about

I need the experience solving the puzzle problems so I can pass them in interviews.

4:51 PM

Your issue is that employers are not able to understand what you can do from your resume?

@Oscar his main issue is that he likes to burn things, and his urges are becoming more difficult to control

^^--- That.

user41796

And we all wondered why he's accused of MOD ABUSE! so often....

You could just put something on your profile "100k+ rep on Stack Overflow in tags X, Y Z"

or something :P

I think, to a certain degree, employers expect to find someone with a "linear" career path. I went from writing business software to analyzing aircraft telemetry data, and employers can't figure out how that fits into their business model.

4:54 PM

That is a point. It is also a pretty cool career

user41796

@RobertHarvey That's true - they have a hard time identifying if their work would be a logical progression for your work

Also: WPF, jQuery, Angular, HTML5 yada yada.

I customize my CVs to the employers I am applying, so I can show them I can add value to their business

That's a full-time career right there.

indeed.

4:55 PM

@Oscar You must do that. Custom cover letters too

[sigh]

pff.

We are a small company. We want our hires to actually care about the business, have it be interesting to them. Cookie cutter resumes mean "I don't actually care about what you do"

"I just want any job"

Here (in Brazil) we do not have this kind of approach :)

yet you make them write bullshit cover letters that only evaluates how well they bullshit.

4:56 PM

that's why you also shouldn't let a recruiter have your resume

I am appplying to germany now and I am going through this process of cover letter and so on.

I think the moment I got this job was when I told my (now) boss that platinum prices were affected by EPA regulations. The rest of the interviewing was formality

@durron597 How do you do that, if you don't know anything about the company? And don't tell me "you didn't do your research." The only way you can know how a company works is by talking to the people who work there.

@RobertHarvey Ask.

Ask who? The receptionist?

4:57 PM

@RobertHarvey google is helpful for this too, even if you just know about their products

user41796

@RobertHarvey reach out through your linked in connections

@RobertHarvey clients of the company

user55340

Check glass door.

One of the people we're interviewing next week emailed us to ask us more about our business. He said "I'm interested in the opportunity and I'd like to learn more about your company" and asked specific questions. Both myself and my boss were impressed, we didn't feel like he was wasting our time.

@GlenH7 This too

You can send people who work there inMessages

and seriously, I don't give a flying crap about your business. making good code doesn't care if their business is widgets or phones or stocks or energy. the business rules are always arcane and they're always meaningless in the end.

4:59 PM

@enderland Also this is good

The idea is that when searching for job postings you'll probably find dozens or hundreds

Think about looking for a job like looking for a significant other

user41796

@Telastyn Shhhh. You're not supposed to say that until after you have the job.

user55340

@Telastyn exactly my feelings.

(making gender assumptions because I'm lazy) would you just date any woman you meet? or are there specific things you look for?

yeh, not so good at biting my tongue.

@MichaelT or subcontractors, people they have subcontracted for etc.

5:00 PM

@durron597 yes, I look for her to be my wife ;)

@enderland you already have a ~~job~~ significant other

hey I have a job too

my point is for when you're ~~unemployed~~ single

@durron597 sure, I look for things, but they're Joel test things. Stuff that actually impacts my job. I don't care what you name your widgets.

stipulations...

5:01 PM

Dating is an apt metaphor. Thanks for the suggestions. I'll incorporate them into my search. But I'm a bit with @Telastyn on this one: why would I be enamored with your company if we haven't even dated yet?

@Telastyn everyone's different, too, I actually care that the product my company makes is a "good" thing for the world (I turned down a job partially because their product only exists because of government regulations which are ~~pointless~~ beneficial)

You shouldn't, but you should know her background before you go on a date :)

@RobertHarvey Just like with a woman, you don't start out enamored. You start out interested/intrigued, so you start putting effort in

If it's the right ~~job~~ woman, you put in more effort. If it's not, you stop putting in effort

and when she start getting boring you dump her and find a new one

@ratchetfreak Ding. this happens with jobs too!

5:03 PM

@enderland quite true.

I love this chatroom

@Telastyn Some companies look for this in their hires. You might LinkedIn message some of their developers and say "Hey I'm interviewing soon and I want to learn more about your culture"

user41796

@enderland ironic in that you could substitute "chatroom" for "job" or "woman" in several of the above comments

I will never do that.

Some companies want their hires to care about the product itself. You (@Telastyn) should not work at these companies.

5:05 PM

because they will lie.

user41796

@Telastyn Not always. Depends upon the level of connection you have

@enderland in soviet Russia, chat room love you

user41796

Buddy of a buddy is likely going to be honest. No connection at all? Meh...

I'll go to glassdoor and check

in my experience, even buddy of buddy are dubious.

user55340

Some places it's a "do you know what you're getting into? It can suck here..."

5:06 PM

@GlenH7 idk, if you find ex employees you are in better luck than current employees

I've done that before

@Telastyn If you ask on LinkedIn "I'm considering interviewing, I really like having a quiet work environment" and then you come in to the interview and it's a really noisy open floor plan then you know you're never going to work with these ~~liars~~ people

I don't see how me asking is going to change that conclusion.

Well, I hope I can hold out for this ideal of "find work that you're suited for, where the employer understands what the words 'mutually-beneficial collaboration' really mean." I hope I can say to a company "This is the place where I genuinely want to work," and not be simply saying it because I'd like to keep a roof over my head.

best of luck with that.

anyways, food. bbiab.

@RobertHarvey Just like in dating, having high standards makes you more attractive, not less

5:08 PM

@durron597 ... to some people, but, generally people you want to date

Yeah; it can also make you look like an arrogant prick, if you don't balance it properly.

@GlenH7 so jobs only go to members of the closed circle who have connections to a company?

@RobertHarvey /me is reminded of the old trope about the most successful playas being arrogant pricks

user41796

@JimmyHoffa No, but they'll go faster based upon internal recommendations

@JimmyHoffa generally the easiest way to get jobs is networking (even or perhaps especially internally)

I mean, imagine that @durron597 lived in LA area, you think @RobertHarvey would have any trouble getting an interview/job with his company?

5:10 PM

@enderland He would, because he hates Java :) (Your point is well taken)

@durron597 the vast majority of people in a company are committing to serious employment risk by responding to any inquiries like this. HR finds out and that person is going to meet the axe asap.

@JimmyHoffa You can ask the HR person yourself!

@JimmyHoffa that's why you talk with ex-employees, not current employees

user41796

@durron597 We all hate Java. Some of us are just better able to tolerate it than others.

@durron597 because HR is the epitome of truthfulness.

@enderland ex-employees pretty much all paint with the same brush: The one that involves them having parted ways for some reason

5:12 PM

@JimmyHoffa If it makes you stand out enough to get an interview, then you've gotten an interview. And if it turns out HR was lying, you don't take the job.

@JimmyHoffa but they have nothing to lose about answering questions truthfully

or at least, a looot less than current employees

I'm with @Telastyn on this, it's a pile of bullshit to expect people to investigate your company in any way other than what's obviously publically available. Beyond that, it's up to the employer to sell the employee if the person's any good - it's the employer who will end up profitting on hiring good people.

most hiring practices are bullshit though. like the CV, that is a test of how good of a salesman someone is. Real useful for sales jobs, but for tech..

Are we talking about interviewing and getting jobs?

oh boy!

The only useful parts of a resume are those that describe actual technical activities completed, and describe technical skills with some sort of relative rating to tell you they're better at SQL than JavaScript or vice versa

The CV and the rest of it is just sales material

@Ampt A bit late to the party, eh?

5:14 PM

@Ampt I heard that being a sockpuppet is a huge plus for employers. Can you confirm/deny?

user55340

Maybe accomplishments, but those can be fluff pieces.

@durron597 well it's great because you get extra hours, not so great because there's no overtime.

@durron597 Hold on while I ask my sock.

@Ampt Wat. Do sockpuppets have a union?

@durron597 well you get your 40 hours, then your sock puppet gets another 40 hours! it's fantastic if you want to work 80 hours in a week

also, no, we tried unionizing to get some gold bond but it didn't go so well

user41796

5:17 PM

@Ampt That's one way to crank up the utilization rate....

@RobertHarvey Sorry, been busy :P

@Ampt What happens when you try to expense it?

they give us to house elves. And I thought muggle feet were dirty....

@Ampt Wow. I don't know if I'd abandon unionizing. Maybe you can get the house elves to strike with you

user41796

@Ampt but no one loves a sock as much as a house elf loves a sock.

5:19 PM

I think it'd be interesting for software devs or IT folk to unionize

probably a terrible idea

@GlenH7 yeah, but try getting him to wash his feet!

but I think the skillsets offered by IT/software folks are grossly undercompensated on the whole considering how many companies are 100% dependent on them

user55340

@enderland debatable indeed.

gah... I feel like unions are so stigmatized right now that it would really make us look even worse

I mean we already have people complaining that we do nothing and add no value

"interesting" is definitely the case, good/bad is very much debatable :)

user55340

5:21 PM

"You may only touch the network cable of your computer if your job includes network engineering." - that's the danger of tech unions.

user41796

@MichaelT I know some network guys who would love a rule like that. Would make their lives much easier

user55340

"Unless you have certifications in Linux, you may not write shell scripts to be run by automated processes"

Unions are like HOA's. A good idea that turned horribly bad, and for the same reasons.

Power corrupts.

Animal House Sock Puppet Show

user55340

A key part of unions is protection of the job description. If people do things outside of that, it means your job may not be as secure - and the union acts to limit that.

user55340

5:27 PM

Part of the problem with tech unions is the best of us are generalized troubleshooters and are tasked with tracking and solving the problem end to end.

@enderland truth. If you broke down their profit and analyzed how much of it came from what, I think you'd find a disturbingly uneven amount of it came from IT, and that the amount coming from IT is a massive multiplier over what their IT costs (on the order of 25 to 50x), so if they paid IT twice as much they'd still be making 12x to 25x off the cost and still be making huge profits and pulling huge salaries and getting a majority of all of that from IT.

user55340

Devops would make a tech union tear out its hair... Hmm.

@JimmyHoffa But then they'd have to lay off all their drinking and golf buddies that just run the test suite and see the developers were telling the truth

@durron597 That was pretty good. Five minutes not altogether wasted, which is more than I can say for most Youtube videos.

@Ampt which is the point. Anti-union folks have successfully ensured that even those who would benefit from it are against it, so people can't act for it without being segregated from the only people who should help them out.

(not saying unions are good, but this is a a mechanism that has been thoughtfully put into place.)

user55340

5:38 PM

What most tech who want unions are just after cross sector collective bargaining. And, btw, the ability to find people to outsource to really hurts that.

@ThomasOwens: Are you still here? I'd like to request a migration of a cross-posted question to CS, to avoid further cross-posts.

Finding Contextual Nodes in a Knowledge Graph

@MichaelT I have no idea if unions in tech would be good or bad, I think it would be a total mixed bag and just a totally different industry than we have in ways we can't predict, but I also think it will absolutely never-ever-ever happen so it's irrelevant to even think about.

@RobertHarvey done

@MichaelT fortunately most large companies have experienced huge pains with outsourcing for the "cheap labor!" route by this point I expect

user55340

It's hard to strike when someone can be brought in got fraction of the price tomorrow.

user41796

5:44 PM

Does anyone know why CS is still in beta?

@ThomasOwens Thanks.

@MichaelT I'd be curious to see the effect this might have on bigger companies

Is it just me, or is this question a contradiction? How can a compiler perform a run-time check?

user41796

@RobertHarvey Made a bit more sense in the first version

user55340

@GlenH7 ?/day is 16. Answer rate is 87%. Compare code review at 35/day and 95%

user41796

5:51 PM

Area51 implied that ?/day is fine. But I don't think they've updated that template in a while

> Have anybody hade a similar idea?

Sheesh.

user55340

It's fine, but it is low compared to other betas graduating.

user55340

Not sure how sticky thier high reps are, or if it's mostly mod run.

Code review will graduate. It's got one of the highest S/N ratios on the network.

user55340

cr is in the pipeline. Cs is not at the moment.

user41796

5:56 PM

makes sense

user55340

Cr has a rather active / sticky high rep population. I am not sure how much happens on cs without a mod acting on it

user41796

Yeah, that would be bad for the site - needs to be community driven not diamond driven

That's a problem happening on Workplace actually :\

@RobertHarvey Halting problem IMO

Active users have become mods and a few others left...

5:59 PM

@enderland When a site becomes too full of duplicate police it's hard for it to stay active. Eventually you begin to exhaust the topic space

user55340

Go to CS. Look at q sorted by newest. Look at vote and answer count. Consider cr had a big push on that prior to graduation.

@RobertHarvey I read that as “will a compiler emit code that performs sanity checks at runtime to decide whether an optimized vectorized code path can be taken, or a safe default path should be used?”

user55340

Click on a closed question. How many close voters?

6:12 PM

@RobertHarvey I answered it fwiw

1

A: Do today's modern compilers perform run time checks?

For a compiler to do run time checks (to find warnings and errors that cannot be found at compile time), it starts to get dangerously close to the Halting Problem. Alan Turing proved in 1936 that a general algorithm to solve the halting problem for all possible program-input pairs cannot exis...

user41796

@MichaelT Wow. Whereas we close off a lot of stuff, it looks like stuff just sits there without any answers or votes or ... anything.

user41796

@durron597 Would you edit the question so it's easier to understand what they're asking? I'll retract my VTC on there if you do. :-)

@GlenH7 It's important to have at least a small core group of dedicated users

user41796

Yes. Over on Eng, the mods are starting to wait on VTC'ing stuff so the community has a chance to step up and do so. And there's already a good core group that's formed.

@GlenH7 I would, but I don't understand what you consider to be the problem with the question.

user41796

6:18 PM

> Do compilers perform run time dependency checks

user41796

Leaves me scratching my head

user41796

Granted, I'm daft. But others are daft too. :-)

user41796

Okay, reading your answer makes the question make more sense.

I edited it a little

@MichaelT I think to succeed in tech unions would have to stop doing that. But I can imagine them being useful without doing that.

6:35 PM

having a submarine at 1:30 am.... feeling hungry

user41796

late night noshing FTW

this is happened for night shift developers

@GlenH7 Can you approve my edit (link above)

user image

user41796

Doh. I should have hit improve, then I could have pushed it through with just one reviewer

@GlenH7 As long as I get my 2 rep :)

Snails pace to 3k

user41796

6:41 PM

@durron597 You would have, yes

user41796

On the "Improve" dialog, there's a little checkbox that says "Original edit was helpful." I'd have to remove the check from that to keep you from getting the repz

@GlenH7 Oh right, I remember that now

user41796

Which I have done on some atrocious edits. But generally I follow the rule of "does it improve the question at all?" and if so, they get their repz

@GlenH7 /me can't wait until I get 3k so I can stop thinking about nickel-and-dime rep

People worry too much about rep than to grab knowledge ..... feeling so... not pointing anyone

user41796

6:46 PM

@durron597 You need to put more answers out there...

@GlenH7 Most questions I want to answer already have quality answers by the time I see the question, and my answer wouldn't add much.

user41796

@JudeNiroshan Many of the regulars in this room use answering questions as a way to learn about something new. Tallying up the imaginary points (repz) just makes it a fun game too.

@JudeNiroshan I just want 3k so I can VTC. I won't care about it anymore after that

user41796

@durron597 Yeah, that can be hard. There will be swaths of time where there's nothing coming through that I'm able to answer or have the time to answer.

user41796

I like digging through old unanswered questions though

user41796

6:48 PM

MichaelT has made some spectacular answers that way. And you know you've got plenty of time to really research them.

@GlenH7 damn.... you're to the fact....

I'm bored now

@durron597 me too :)

user41796

10k is fun. Still tidbits of garbage from the NPR days that need to be nuked. Put the final VTD in place and see it turn a pretty shade

Hm. How do I answer a question when my entire answer is the wikipedia entry?

I don't even know how I would "add effort" to that one

user41796

@durron597 Just explain a bit more how applying an r-tree to that domain simplifies answering the questions they want to ask about the data

6:58 PM

@GlenH7 Well, at least the OP was happy with it.

« first day (1637 days earlier) ← previous day next day → last day (3356 days later) »

Transcript for

Feb '1526

The Whiteboard

General Discussion for softwareengineering.stackexchange.com P...

coffee-day horror-stories monads my-code-is-compiling scotch the-other-monitor

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile