« first day (1686 days earlier)      last day (3307 days later) » 
00:00 - 19:0019:00 - 00:00

7:00 PM
I operate on trees?
that's going on the CV.
 
I have less disdain than I used to for people capable of pooping out a mobile app after reading some tutorials. I've worked with/interviewed too many that were not even capable of that.
 
user41796
@Telastyn That's true. I just happen to be cleaning up some atrocious code "written" by someone who didn't know what they were doing
 
user41796
When I only want futures and you return a current value, you've failed the requirements. Just saying.
 
my condolences.
 
lol @ pooping out a mobile app.
 
user41796
7:06 PM
It's ... frustrating. Fortunately, the person isn't on our team anymore. And this is the last major section of code they wrote. It sucks that I had to clean up the other major section of code they wrote too.
 
@GlenH7 Ballpark value on average lines of code per method?
 
that is frustrating. at least they're no longer employed by you.
 
user41796
@durron597 negative value
 
I have someone I have to clean up after, but they're still employed and have a higher position than I do.
 
user41796
The entire time that person was on the team, they were a net load on us and drug down productivity
 
7:10 PM
at least this guy is only sometimes in need of cleanup, last job was far worse in that regard.
people suck.
 
user41796
That includes a longer-than-normal ramp up time, and just a load of pure schlock that the person wrote.
 
@Telastyn I was once "given" a developer to help me reach a project deadline. I told the project manager this person's contributions would be a net negative. For the next 9 months I spent hours each day helping the dev write code and hours in the evening redoing their work. Good times.
 
@GlenH7 Ballpark AMOUNT of lines per code per method?
 
user41796
@durron597 Sorry, I got stuck on value
 
0
Q: Pathfinding in a 3d world with obstacles?

SolrikWhat are some methods for pathfinding in a 3d world? Object X needs to move from Point A to Point B but it might have to avoid things like asteroids, planets or gas pockets that makes the object slower. The program uses an octtree and is not "tile" or "grid" based.

 
7:12 PM
@MetaFight I would have documented the time spent every day for a month, then made pretty excel graphs and showed it to the project manager.
 
^^ -- This question is the equivalent of icanhazcodez on Stack Overflow. Do we just summarily close these, or would the site benefit from a decent, comprehensive answer?
 
@RobertHarvey Well, I just spent my last close vote on it, that tells you what I think of it.
 
user41796
@durron597 It varied. Generally in mid double digits. Biggest issue was that class constructs didn't even come close to respecting the domain and failed to respect the basics of SOLID principles.
 
@MetaFight - I had a developer like that once. I assigned them some non-critical bullshit work until I could get them fired.
 
user41796
@RobertHarvey burn with prejudice
 
7:14 PM
@durron597 Oh, everyone knew it was happening. I didn't need proof. They just didn't have the heart to fire the guy... so they dumped the problem on me and the project.
 
@MetaFight So show the proof to someone who had profit sharing incentives.
 
user41796
@durron597 Remember that there are significantly different cultural norms for his locale
 
@Telastyn That's what I did the second and third time it happened to me. I assigned the two developers do some research. As a starting point I gave then an MSDN article. Two weeks later both developers, working in isolation, sent me word documents with the very same MSDN article pasted in it.
 
facepalm
 
@MetaFight That's hilarious.
 
7:16 PM
yeah, I couldn't help but laugh.
 
see, that's why I insist on interviewing people
at least then I have nobody to blame but myself.
 
These were folks that were hired without being interviewed?
 
They were interviewed, but by somebody who didn't know what they were doing.
 
mine was hired before I got there.
 
And, again, instead of firing them when it became obvious they were worthless, the manager just moved them from project to project.
 
7:18 PM
but I find that others have faaaaaaar lower standard for what a competent developer looks like.
 
lol
 
That reminds me of a question I wanted to ask. How do you guys prepare for interviews? Let's say the coding screen is out of the way, and you did well. What do you do to bone up for the next one?
 
nothing.
 
This is in Formula 1, where we're supposed to be "efficient machines."
 
I might look at glassdoor to get targeted questions about their workplace culture.
 
7:19 PM
Firing employees is very difficult in most companies
 
user41796
@RobertHarvey Definitely study up on the company some more, so you can ask more detailed questions during the interview.
 
@enderland Some more so than others. It's difficult to fire someone in direct proportion to how worried the company is about the legal ramifications.
 
user41796
Later rounds tend to be more of the softer stuff, which is where behavioral based interviewing tends to come out. Start thinking through safe war stories that can be shared to highlight your abilities
 
@RobertHarvey What Glen said imo. There's not much else you could do.
 
7:23 PM
I might get a haircut if warranted
 
user41796
haircut; have clothes pressed; try to remember to shower that day...
 
shave
 
Sheesh. I didn't just emerge from a cave, y'know. :P
 
user41796
It's also a good time to reach out to 2nd and 3rd connections through LinkedIn to see what sort of recon you can do
 
well, some of us did.
:D
but I also don't get a lot of offers, so my advice might be poor.
I might be back up to 1:10 now.
 
7:27 PM
Is that your ratio of offers to applications? That seems respectable.
 
user41796
IMO, cultural fit becomes a big factor for later rounds of interviewing. They've determined you can likely do the job. Now they need to figure out if you can play nicely.
 
offers to interviews.
that's also probably interviews to applications now too.
cultural fit...
/me spits.
 
"Cultural fit" is just code for "Can you play well in our sandbox with others?"
 
"Cultural fit" is code for discriminatory practices.
 
Well, there's that too.
 
7:30 PM
I've yet to hear a convincing example of "not being a cultural fit" that wasn't just a euphemism for someone being mean or immoral or some other obviously unacceptable to everyone kind of thing
 
user41796
"Unfair discrimination" is a problem
 
user41796
But everyone makes discriminatory decisions. Please also note that discrimination really just means "to decide". But there's a whole lot of social context wrapped around that term
 
true
 
@GlenH7 I was about to point that out.
 
I just find that too many interviewers care about things that have no impact on ability to get shit done.
 
user41796
7:31 PM
If I discriminate against someone because of their color of hair, for example, that's completely unfair
 
people seem to use "discrimination" to mean "deciding whether to hire based on who someone is, not on what they can do or how they behave"
 
but that's the sort of comment that usually results in me being told to shutup.
 
user41796
I may not like blue haired people but it really shouldn't have an effect on ability to get along or do the job
 
@Ixrec Yes, and how many people are going to willfully admit that in an interview? Apparently, a significant percentage. I asked the college career counselor about their "diversity" interview question, and she said that there are people who will readily admit that they are racist.
Asking the question eliminates those people.
 
I read an article about that a month or two ago... let me see if I can find it
 
7:32 PM
@RobertHarvey I've read the DailyWTF, so I've heard of many such examples
 
user41796
But if I decide not to extend an offer because of some obnoxious affectation (which ultimately demonstrates low emotional intelligence) then I'm not as convinced that's unfair discrimination
 
user41796
"Cultural fit" is another way of figuring out guessing at if you're going to end up hating someone's guts after working with them for a few months.
 
user41796
For example - this persons code I'm cleaning up. If the team had gone out to lunch with that person, I think we would have seen a number of red flags that would have forewarned us of the cultural fit issues we ran into.
 
user41796
Their approach to coding (and life in general....) didn't align with how the team views the code.
 
7:37 PM
could you be more specific?
 
meh, if the person writes WTF code, you don't have a cultural fit problem. You have a WTF problem.
 
user41796
And to be fair to the recently departed (pun only) - that person had a difficult and frustrating time working with us. They chose to voluntarily leave. We weren't trying to chase them out, but they were very much unhappy in what they were doing. So there's a balancing act to that perspective
 
user41796
@Ixrec Nope. I'm staying intentionally vague in order to protect the recently departed's identity.
 
Did that person go through a code screening?
 
user41796
All I can present is my side of the story. It's unfair to them since they can't provide theirs.
 
user41796
7:39 PM
@RobertHarvey Yes, but we glom everything together in a batch of interviews. Probably something we need to revisit from a process perspective
 
user41796
And yes, some flags popped up during the screening. I have since modified my screening tests accordingly.
 
The first time I was asked to help hire someone, I didn't know about code screenings. This fellow said he'd written a hundred VB programs. What he didn't say was that they were all the same program with the same set of global variables, slightly modified for each task.
 
:( My autogenerated code doesn't implement clone.
 
Needless to say, he didn't work out. Amazingly, my boss asked me to help hire the next two anyway, who still work there today.
Those two got a code screen, and much better questions from me.
Experience is what you get right after you need it.
 
Speaking of interviews, I have mixed feelings about interviewing interns and co-ops.
 
7:43 PM
3 hours ago, by MichaelT
Intern jobs are in part a n month interview with a too broad question that can't be answered in an hour.
 
@ThomasOwens hire smart people
 
Yeah. But when they don't have good work, it's so disappointing. More than once, I've brought in amazing people. But then things beyond everyone's control happens. The contract falls through or gets delayed. Then, they get stuck working on crap for 3-6 months.
 
@GlenH7 I'm curious, what specifically did you change?
 
@ThomasOwens I had an intern that worked with me on my last project. She did great work, but mentoring really slows you down for your own work.
 
user41796
@ThomasOwens Gotta get buy-in from management ahead of time to prevent that.
 
7:45 PM
@GlenH7 I'm surprised that's not something you would want public. It's pretty general
 
user41796
<--- Is paranoid.
 
@GlenH7 Yeah. But if you hire a co-op or intern many months out...
 
Why do you think that particular change would have caught this person when the previous kind would not have?
 
user41796
@ThomasOwens In general, management has to commit to having an attractive internship program or colleges will move the company to the "wrong list" for promoting the positions being available
 
Really? A vibrant intern program is a prerequisite for college endorsement?
Very few companies these days have good intern programs in my experience. Except for Employer^^, who preferred to hire intern-level people because they cost less money.
 
7:48 PM
@GlenH7 We need to be careful with how we advertise the positions.
 
user41796
@RobertHarvey It was for my undergrad Uni. They had an A tier and B and C tiers for internships
 
But we look for certain skills and interests. We do make sure that everyone's clear that things could happen that change priorities or available work.
 
11
A: What are some best practices for creating an internship opportunity?

enderlandMust-haves (roughly ordered): Adequate management (on part of day-to-day managers) is critical. Many if not most interns will have minimal work experience prior to this assignment, yet, oftentimes, the tendency of managers is to treat them the same as a regular employee. "Here's work, come back...

Colleges have a huge interest in companies having vibrant intern programs and hiring from them
 
user55340
@Ixrec I'm a big company type. I enjoy making working processes. Build tools, bug trackers, deployments. I'm not a good cultural fit for smaller seat of our pants types. Cowboys and Heroes wouldn't be a good fit for where I like to work.
 
psr
I got hired once for a summer internship with a senior developer, who got laid off shortly before the internship began. They had no other work available, but had to honor the agreement with my university. They didn't bother finding me one thing to do. I still got paid, but not a good experience.
 
7:54 PM
@MichaelT Good point. I'm the same way, which is why I didn't apply to any startups.
 
user55340
(Reason Employer^^ can't hire full time locals is they've hired them as interns)
 
@MichaelT perfect irony there. ha
 
user41796
@psr Did they still make you show up at the workplace?
 
psr
@GlenH7 Yep. And I was a few minutes late a couple of times over the course of the summer, and they were very displeased. I had to begin doing my nothing on schedule.
 
user41796
The bitterness of the irony there is palpable
 
user41796
7:59 PM
If they ever wrote you up, that would be worth framing
 
psr
They were a mess. I felt bad for the FTEs. There was a high-up manager that would walk through the building, and if something he saw or heard displeased him, he would fire people on the spot. There would be a stampede running away like animals in a forest fire, then he would walk by. I stayed put because I was nearly the most un-firable (and the stakes were much lower for me). It was weird and sad.
 
user41796
"Thank you for demonstrating why I never want to work here."
 
8:34 PM
I don't know what to do...my generated code doesn't implement clone(), but I submitted a request to use the software that allows me to generate it. Fortunately, all of the licenses are good and most of the dependencies are Apache Commons stuff which has been approved. But if that doesn't work, is serialize and unserialize to clone really my best option?
 
it's not a horrific option in practice.
 
It's much slower than clone.
 
I mean, not any more horrific than needing clone in the first place.
 
user55340
Could serialize through xml
 
8:37 PM
@MichaelT Yeah. This is a JAXB generated code, so I would.
 
user55340
One deep clone approach is to send it to xml stream and back out.
 
Yeah. That's exactly what I was thinking would be my option.
Also, projects that don't provide a binary download and expect me to use Maven suck.
Please do releases and provide each release as a tarball and/or zip.
Because otherwise, I need to do that anyway, and I'm not going to.
 
user55340
Dependencies are a problem then. It's easy to get the binaries though.
 
user55340
(You can easily down load a jar from maven without maven)
 
@MichaelT Package the dependencies, too.
 
user55340
8:41 PM
That involved licensing and distribution fun for them.
 
It depends on the licensing of the dependencies, doesn't it?
 
user55340
Yep.
 
Although this particular project...has no dependencies.
 
user55340
And if they can distribute them.
 
Well, it's an extension to JAXB. So it depends on JAXB. But that's part of the JDK. It has no other dependencies.
 
8:43 PM
Out of CVs on both SO and programmers now. One of those days...
 
user55340
And means that you may end up with multiple versions in your projects.
 
@MichaelT That's actually a good thing.
At least for me, knowing exactly what version I have and not dealing with Maven is a good thing. It's less of a headache. Every new version needs rescan and reapproval.
 
user55340
Sealed jar collisions are not fun
 
Alternatively, you don't need to provide dependencies if you list them out in a README.
 
user55340
It is in a readme- Pom.xml
 
8:45 PM
That's still Maven. Do you know how hard it is to get things in the door?
Simplest case, in a non-security environment, requires license review, virus scan, and vulnerability analysis. For every file of every version.
 
user55340
Pom is very human readable.
 
user55340
@ThomasOwens no dependencies tag, no dependencies.
 
Yeah. No dependencies in mine. But I need to use Maven. But we can't always download in projects, so I need to download and build this one version, run the scans, submit for license approval, and then upload it to the repository of approved software.
 
user55340
8:48 PM
Check parent though. <artifactId>jaxb-plugin-parent</artifactId>
 
Do deleted on hold (as opposed to closed) questions automatically get one undelete vote?
I'm looking at undelete votes on SO and some of those questions are godawful and unredeemable. Why in the world would they have undelete votes?
 
@MichaelT See. Why Maven is awful in my environment. Also, no. Just JAXB and JUnit as dependencies, so nothing special.
 
user55340
There are some on SO who believe all should be on the site. Never delete anything with some modicum of utility, even if better found elsewhere.
 
But what would help me is just a text file of dependencies and version numbers and a ZIP or TAR of the prebuilt application. Including dependencies in the ZIP or TAR would be good, if it's allowed. Let me manage my versions and walk it through security.
Honestly, in 90% of the world, tools like Maven are great.
But when you're dealing with a secure environment and very tightly managed software, down to the version of every single dependency, you lose a lot of control.
 
user55340
Yep. I know, in that env now, without maven. Trying to get it in.
 
8:53 PM
Trying to add Maven?
 
user55340
Yep.
 
Although part of the problem is that a lot of my builds are on machines not connected to the Internet.
So it becomes a manual process anyway.
 
user55340
There are ways to get maven to use a local repo and not download the Internet.
 
Yeah, but creating the local repository is a pain.
Although, honestly, maybe we should look into it....
 
user55340
So is downloading all the dependencies.
 
user55340
8:54 PM
Look into nexus for maven.
 
How well does Maven play with things not Java?
C, C++, etc.
 
user55340
 
user55340
Reasonably. It's a matter of the build plugin.
 
user55340
Employer^^ used nexus because our CI machine wasn't able to access the Internet. Nor were our interns.
 
Wait...stop. Your interns couldn't access the Internet?
 
user55340
8:57 PM
Nope.
 
user55340
Want java docs? There is a printer.
 
I should look at tools like Maven.
There could be something better for our purposes.
 
user55340
Or gradle or ivy.
 
I wonder if it also works for things that aren't part of a build. For example, installers. Let's say that a certain FOSS tool or non-corporate supported tool is desired, we need a way to approve and maintain approved versions of those.
Do you think a tool like Nexus would work for individuals (not all software engineers, either), to obtain binaries to execute as installers?
 
9:15 PM
Part of the reason I resist learning Haskell (and I'm serious about this) is feeling like my hands are going to be tied behind my back returning to a less expressive language.
 
hmmm new job has recurring 7am meeting on wednesdays.......
 
Don't know if this question is salvageable, but he made all of the changes I requested, and I voted to reopen:
-3
Q: Pathfinding in a 3d world with obstacles?

SolrikWhat are some methods for pathfinding in a 3d world? Object X needs to move from Point A to Point B but it might have to avoid things like asteroids, planets or gas pockets that makes the object slower. The program uses an octtree and is not "tile" or "grid" based. E.g.: What kind of pathfinding...

 
Similarly, I can never get enthused about learning new programming languages because I'll never be able to use them for anything useful, unless I spend weeks in the FFI making it useful first
@RobertHarvey imo it's successfully achieved "neutral" status now, i.e. neither worthy of an upvote or downvote, but definitely answerable (and answered) so I'll VTR too
 
The purpose of these sites is to get useful answers to actionable problems. The problem posed in your question (and your corresponding observations) are neither actionable nor useful. — Robert Harvey 51 secs ago
 
9:39 PM
The nice thing about Haskell is that even though it's very expressive, you still can't do much in it due to lack of (practical) library support (at least when I played with it).
Without a whole lot of pain/annoyance.
nice for Robert Harvey's concerns - not nice for Haskell in general...
 
psr
10:07 PM
@Telastyn I found a theorem prover for sequent calculus right away. hackage.haskell.org/package/pesca. If I knew what sequent calculus was that might be very practical.
 
10:53 PM
@psr seriously? That is highly unusual. Sequent calculus isn't supposed to exist for at least another hundred and fifty years, I'm not sure how they even backported it's proofs into LaTex...
@psr the fact that it's not even hard to find people with stories like this really makes me wonder where people get their endless optimism in the face of such evidence
 
user55340
11:22 PM
@ThomasOwens It can. An artifact is something and that has a version. We deployed our .war files to the nexus repository also - there's nothing saying that one can't deployed traditional .jar with main methods with the intent of pulling the latest one for some other build process.
 
user55340
@ThomasOwens It could very well do so. I would be tempted to investigate gradle more as it is a full groovy scripting language build system rather than xml - it would allow you a bit more flexibility than maven or ant (ivy) would provide.
 
psr
@JimmyHoffa Someone had it try to prove that backporting to LaTex was impossible and it actually disproved by counter example. Oddly, that's the same way they discovered time travel.
(Assuming I remember f a z ( rot x f 13) (x->vvl) _ (a:_,_) _ -> a Prove rNatProof p1 ++++ "}" DisjIr (map ipso latorum) ++++ -<> correctly)
 
this question challenges us to put some votes on it: programmers.stackexchange.com/questions/279355/…
 
psr
The quality filter needs to check for "cus", or at least "cus in school".
 
00:00 - 19:0019:00 - 00:00

« first day (1686 days earlier)      last day (3307 days later) »