|
sarehu posted:That's actually a very easy question for real mathematicians, just not for quiche eaters that took math classes without doing any math in their spare time.
|
#
?
Sep 22, 2015 05:50
|
|
|
|
| # ? May 18, 2024 00:29 |
|
|
Tao Jones posted:What I've found works well for setting up a phone interview is to give two times that you're available and let them choose between those. That gives the impression that you're a Busy Professional With Stuff Going On (even when that's not the case) and prevents any "analysis paralysis" on the part of whoever's populating a schedule. If I called a doctor's office and tried to make an appointment, it would be weird if the receptionist was like "oh the doctor's schedule is pretty open, just come in whenever and they'll see you" not only because I'd wonder what kind of doctor has complete availability, but also because then I'd have to wring my hands and figure out what time out of all of the times I have available is the best, which probably will lead to my making no decision at all. Well now that they drug their feet people who can use a goddamn phone have already scheduled with me so that's no longer an issue. Analysis paralysis sucks anyway.
|
|
#
?
Sep 22, 2015 05:51
|
|
|
I'd like some career advice please. I will try and be as concise as possible. I'm 33, spent most of my 20s in manual testing/mobile telecomms. Eventually ended up in an automated testing role where I was doing a lot of programming and creating tools. Realised I liked it a lot more than testing, started doing an Open University degree specialising in software engineering. Was made redundant from automated testing role (company was shut down), and decided to try and get a dev job instead of another testing job. Managed to get a small company where I knew a couple of people to take me on as a junior .net software dev - took a pay cut from £33k to £25k to do so. While I've been here I've been working on two projects - an ASP.NET MVC app we've been developing from scratch, and a pretty shityy Winforms app with a horrible codebase I inherited. From very early on in the role I've been doing what I consider to be non-junior stuff on the Winforms project - dealing directly with customers, doing requirement analysis, design, test, implementation, roll out and test - the whole life cycle basically (there are only 2 of us in the dev team - the company does IT support with a bit of dev on the side). For the last few months I've been solely on the Winforms project and basically running it - my manager just joins the weekly conference for an update, and sometimes doesn't even bother doing that - just lets me chair it and read the minutes. After 6 months I was give the promised pay review but it amounted to them saying they couldn't afford to give me one. That was the last time pay was mentioned until a few months ago I was taken aside and spoken to because I seemed a bit demotivated. I told them I felt I was being underpaid given the work I did - they agreed to immediately remove the Junior from my job title and review my pay at the beginning of September. Meanwhile I decided to check out the job market anyway. Applied for and landed another job - they offered to take me on at £30k, with a review to a minimum of £32k after my 6 month probation, and 37.5 hours a week (I forgot to mention, the week here is 42.5 hours). Also, the September pay rise wasn't mentioned again. The company is a little larger - 16 employees as opposed to 9, and their whole business is development. I gave my notice - citing low pay, long hours, not enough variety in the work. They offered me £27k to stay (saying they were planning to give me that anyway), I turned it down. At this point I have 2 weeks notice left. Current employer has been advertising my role, asking for applcations with 2-3 years experience, offering £25k-30k. They have been getting very few applicants and a couple of recruiters have told them that this is unrealistic. I think they have now realised that, because this morning they offered me £32k and a reduction to 37.5 hours a week to stay, plus rebalancing of worksloads between the 2 of us so I get a bit more variety. I think I already know my answer, but I'd really appreciate some opinions on whether I should take it. On the one hand - I do quite like it here, I'd be on 32 straight away instead of in 6 months (and I have many troubles at the moment), and I am somewhat terrified of starting a new job. On the other - I dont know when they would review my pay again (probably never lol), I am pretty bored with the work and it'd be nice to work somehwere a bit bigger, even if it's not that much bigger. And of course there's the ethical thing - I've accepted the other offer now. What say you, goons? chippy fucked around with this message at 12:19 on Sep 22, 2015 |
|
#
?
Sep 22, 2015 12:16
|
|
|
I feel like you know the answer, but just typed it out in order to give yourself clarity of mind. Which is fine, but there you go. Take the new job, stagnation is the second-greatest mind-killer after fear. It's time to grow. Re: math and the n! thing. I've been trying to expand my knowledge of math, particularly related to programming. I've worked my way through a pretty big portion of the Project Euler archives, and since there's over five hundred challenges I'll be busy for awhile, but I was hoping you guys had some other resources to use alongside it. I've learned a LOT, both about math and programming, and making them work together, however a more focused series of challenges that expand on the previous ones would be preferable. Any ideas?
|
|
#
?
Sep 22, 2015 13:04
|
|
|
sarehu posted:That's actually a very easy question for real mathematicians, just not for quiche eaters that took math classes without doing any math in their spare time. Supreme-laffo at non-vegans who don't eat quiche.
|
|
#
?
Sep 22, 2015 13:14
|
|
|
chippy posted:What say you, goons? Hmm, jerked you around until you got fed up and told them you were done, then suddenly found the time to actually review your pay along with money they previously said they didn't have? In the 'States this is where I'd say "they just want you to stick around while they look for someone cheaper" but I'm not sure what laws are in place in the UK to prevent that kind of nonsense. I will say that boredom is a terrible thing to deal with as a programmer and if it were me (and not knowing just how tight your financials are) I would weigh that pretty heavily. Also I would want to get away from WinForms, and for the same reason as I would want to avoid boredom: if you're not learning and staying reasonably current, you're going to have a harder time finding jobs later on. At least, jobs that aren't just a bunch of legacy maintenance (which if that's your thing, fine, but that bores the poo poo out of me over a long period of time). As you said though, it sounds like you already know your own mind.
|
|
#
?
Sep 22, 2015 14:51
|
|
|
chippy posted:
You'll never get back to a good relationship with them. You know they did not value you properly, and are only making the offer out of panic. You'll also get bored soon of the WinForms crap, and as Che said, it's not worth knowing anyways. Take the new job because it will be better for you in every aspect, and don't let your old job guilt you over things they had many opportunities to fix.
|
|
#
?
Sep 22, 2015 16:21
|
|
|
fritz posted:Supreme-laffo at non-vegans who don't eat quiche. Also, vegans.
|
|
#
?
Sep 22, 2015 18:26
|
|
|
Yeah as someone who's taken a counter-offer after putting in their two weeks: don't take a counter-offer after putting in your two weeks.
|
|
#
?
Sep 22, 2015 19:09
|
|
|
So I just had an interview where a guy asked me to do $CTCI_THING and I said "I just did it, here's the code" so he upped the ante a few times until we did poo poo like reversing a linked list without making a new one (pointer shufflin') and such. Also one of the before-the-interview questions, which I did on paper and scanned (lol) was to check if strings were anagrams. I just made two character arrays, one for each string, sorted, and used charArray1.SequenceEqual(charArray2); instead of doing some shenanigans with buckets for each char in the string or hashing. Interviewer said he liked my method better even though it's slower since it's concise, clear, and easier to understand.  What I mean is gently caress the imposter syndrome, keep trying, yadda yadda.
|
|
#
?
Sep 22, 2015 21:42
|
|
|
Space Whale posted:Interviewer said he liked my method better even though it's slower since it's concise, clear, and easier to understand. LOL. Should've said you used bucket sort. edit: Or sperg out about hash table worst case running time, sheesh. sarehu fucked around with this message at 21:51 on Sep 22, 2015 |
|
#
?
Sep 22, 2015 21:49
|
|
|
sarehu posted:LOL. Should've said you used bucket sort. Well, see, I did, and he said "yeah but in practice debuggable simple stuff is better than something more complicated but more performant, in my opinion."
|
|
#
?
Sep 22, 2015 23:24
|
|
|
Space Whale posted:Well, see, I did, and he said "yeah but in practice debuggable simple stuff is better than something more complicated but more performant, in my opinion." HA! Don't work there then, there are likely a ton of massive performance issues that need to be resolved and causing giant customer facing fires every day.
|
|
#
?
Sep 22, 2015 23:27
|
|
|
Skandranon posted:HA! Don't work there then, there are likely a ton of massive performance issues that need to be resolved and causing giant customer facing fires every day. I'm pretty sure a 30 year old company that makes academic performance evaluation/grading software needs O(constant) time string anagram comparisons. Also they know how to op poo poo. They have automated essay grading. Didn't divulge too much in the interview but holy poo poo.
|
|
#
?
Sep 22, 2015 23:38
|
|
|
Space Whale posted:Also since I just wasted my time beating my head on a wall for lack of remembering nCr, nPr, and other combinatorial, discrete things, and bombed a codility interview, please tell me what the gently caress I screwed up. Mind you I've been a paid programmer for three years and haven't touched or even seen this kind of a problem in that time. I have a feeling this is one of those cases where you shouldn't beat yourself up over something outside your control. Most people learn what's important for them to know and combinatorics (while fantastic for understanding upper bounds for CS and understanding the structure of multiple outcomes) isn't relevant when you don't focus on formal analysis of existing procedures. Which honestly is a lot of jobs, especially at the lower levels where you don't have the confidence/knowledge to critique existing ways of doing things.
|
|
#
?
Sep 23, 2015 03:28
|
|
|
Skandranon posted:HA! Don't work there then, there are likely a ton of massive performance issues that need to be resolved and causing giant customer facing fires every day. To be fair that can actually be true quite a lot. Sometimes you can't reduce the complexity very far and the performance difference isn't all that big. In other cases it's a matter of knowing that n will never, ever be large. No program is ever free of bugs so sometimes slightly worse performance is more important than something simple and unlikely to break. It's hypothetical but if you have an O(n squared) algorithm that never fails, ever and an O(n) one that occasionally does n squared will sometimes be the one you use. If n never gets over 100 then who cares? That's 10,000 worst case. Unless you're doing huge calculations that isn't likely to cause problems. Plus, far as I can tell, a lot of programming is "OK this works and the performance is acceptable? Use it and work on something else." You can always, always, always optimize but the question is, should you? The performance increase may or may not be worth it and the simple fact is that easy to read and debug code is extremely important. If you don't look at that code for ten years and then quit the company can the new guy figure out what that does? If yes, then awesome. If no, it doesn't matter how good it is you might very well end up paying a programmer to rewrite it.
|
|
#
?
Sep 23, 2015 03:41
|
|
|
It's not necessarily true that you can "always, always, always optimize", but that's mostly in the case where you're doing insane low-level micro-optimizations (meaning optimizations beyond the notice of big-O) like using big memory pools and custom memory layouts so you reduce cache misses. Sometimes you really have to structure a program around an optimization, but it only really comes up if you're reasonably sure from the start that it's necessary. It's unlikely to come up in most programs you write unless you're the lead engine programmer on Unreal 4 or something I guess. I suppose it's technically true that you can always optimize later, but there are some cases where the important optimizations are so intertwined into the core of the program that the refactoring would essentially just be a total rewrite. Still, it's not a thing you should be doing unless you have the experience to know ahead of time that it'll be a problem in the future if you don't, and enough knowledge to mitigate the maintenance drawbacks.
|
|
#
?
Sep 23, 2015 03:50
|
|
|
The problem he was discussing is easy to optimize and they were both in fact aware of it. It also leads to significant savings. To prefer not to do that is like preferring Selection Sort over Quicksort because "Quicksort is hard, I prefer it to be easier to understand and debug".
|
|
#
?
Sep 23, 2015 05:00
|
|
|
I should probably point out I did this on pen and paper, not even a text editor.
|
|
#
?
Sep 23, 2015 05:31
|
|
|
Space Whale posted:I should probably point out I did this on pen and paper, not even a text editor. Sorry, not meaning to sound like criticizing you, but the interviewer. Knowing the better solution but then saying he likes the worse one is just dumb.
|
|
#
?
Sep 23, 2015 06:12
|
|
|
Skandranon posted:Sorry, not meaning to sound like criticizing you, but the interviewer. Knowing the better solution but then saying he likes the worse one is just dumb. Also how long of a string would you ever anagram-test? Seriously, now. Besides, if it was ever huge strings wouldn't you go whole hog optimizing it and call a C/C++ DLL anyway?
|
|
#
?
Sep 23, 2015 06:35
|
|
|
Space Whale posted:Also how long of a string would you ever anagram-test? Then the question (well in the real world anyway) would be how often you could possibly get very long anagrams. In that case it would probably be very unlikely it would actually pass. What are the chances of a million character string ever being an anagram? While there are probably situations where that is true I really can't see that happening often.
|
|
#
?
Sep 23, 2015 08:01
|
|
|
Thanks for your input guys, You're right, I had pretty much made my mind up, but the fact that no-one has made a case for staying in my current job has helped reinforce it as the right decision.
|
|
#
?
Sep 23, 2015 08:58
|
|
|
ToxicSlurpee posted:To be fair that can actually be true quite a lot. Sometimes you can't reduce the complexity very far and the performance difference isn't all that big. In other cases it's a matter of knowing that n will never, ever be large. No program is ever free of bugs so sometimes slightly worse performance is more important than something simple and unlikely to break. Your profiler should be your guide. Just write something that works (and isn't insane), profile it with real data, and optimize what shows up in the profiler.
|
|
#
?
Sep 23, 2015 12:39
|
|
|
ToxicSlurpee posted:Then the question (well in the real world anyway) would be how often you could possibly get very long anagrams. In that case it would probably be very unlikely it would actually pass. What are the chances of a million character string ever being an anagram? While there are probably situations where that is true I really can't see that happening often. Gene sequence processing, maybe? The concept of an anagram would be a little more complicated there but there'd be real memory savings in a bucket counter approach because there are a comparatively tiny number of protein encodings.
|
|
#
?
Sep 23, 2015 15:00
|
|
|
Skandranon posted:The problem he was discussing is easy to optimize and they were both in fact aware of it. It also leads to significant savings. To prefer not to do that is like preferring Selection Sort over Quicksort because "Quicksort is hard, I prefer it to be easier to understand and debug". No it's not, because your example is an n/log(n) factor and his is a log(n) factor. Not only that, it's a log(n) factor that realistically fits inside a cache line, not some memory-hopping log(n) factor. It's probably faster than what your implementation that uses a hash table would be. Guess what, in real software you sometimes say "This O(n^2) solution is good enough." Get a job and put some theory into practice sometime. The interviewer was right, the simplicity far outweighs some imaginary log(n) performance benefit.
|
|
#
?
Sep 23, 2015 21:32
|
|
|
sarehu posted:No it's not, because your example is an n/log(n) factor and his is a log(n) factor. Not only that, it's a log(n) factor that realistically fits inside a cache line, not some memory-hopping log(n) factor. It's probably faster than what your implementation that uses a hash table would be. I didn't post a solution, I was commenting about deliberately taking the slower one. The only time taking the slower one is when it's not worth the time to invest in making it better, but if you already HAVE both in front of you, it's dumb not to use the better one. Space Whale posted:Interviewer said he liked my method better even though it's slower since it's concise, clear, and easier to understand. Also, I have a job, and one I'm quite happy with, so I don't need to come on here and be a dick to everyone because I hate my job.
|
|
#
?
Sep 23, 2015 21:36
|
|
|
You don't have both solutions. Thinking of the solutions is not the hard part. The version that sorts is less code to write and maintain and is really hard to gently caress up. You don't have to write any loops or anything. Also I'm not sure why you think I hate my job. I'm pretty sure you're wrong on that count. Sorry I disrupted your delusions of grandeur as you sat on your high little horse trash-talking some guy you've never met for a perfectly reasonable opinion. (NOT.)
|
|
#
?
Sep 23, 2015 21:46
|
|
|
sarehu posted:You don't have both solutions. Thinking of the solutions is not the hard part. The version that sorts is less code to write and maintain and is really hard to gently caress up. You don't have to write any loops or anything. I just assumed you hate your job because that's the only rational excuse I could think of for why you're such a horrible person. Must just be you then.
|
|
#
?
Sep 23, 2015 21:49
|
|
|
Wow the newb-help thread is picante today
|
|
#
?
Sep 23, 2015 21:57
|
|
|
Wait you quoted me but you say Sarehu hates his job?  Also I had a technical skype interview with some coding and now I'm being brought in for a technical interview - what should I expect besides "whiteboarding"? Anyway I hope I have a job again soon  Necc0 posted:Wow the newb-help thread is picante today Olé!
|
|
#
?
Sep 23, 2015 22:09
|
|
|
Going to be inbetween projects at my current full time job, i'm basically been given the opportunity to study for some certifications and take them while getting paid. I was thinking about getting into web development. Does anyone here have any comments on these certifications? I'm getting a free ride here as the company seems to be investing in me instead of laying me off. I was looking at MCSD Web Apps. Generally just looking for feedback on certifications, if it helped you land a job, higher salary, was it not worth it at all for you, etc. EDIT: I have a strong background in C# and WPF - coming to realization that WPF seems to be pretty niche and not in as much demand as I thought.
|
|
#
?
Sep 24, 2015 02:44
|
|
|
In the programming world, certs are generally useless. I've heard of some being useful in specific contexts, and also of a negative bias against them from other employers. Many competent software engineers seem to view them on a resume as a signal that the applicant is clueless.
|
|
#
?
Sep 24, 2015 03:32
|
|
|
I think it mostly depends whether you're doing government work or not. I know a few people employed by either the Department of Energy or DOD/Military as software engineers that needed to get certs to be hired, or were hired on a conditional basis that they passed a certain certification exam.
|
|
#
?
Sep 24, 2015 03:56
|
|
|
authwiggidywack posted:Going to be inbetween projects at my current full time job, i'm basically been given the opportunity to study for some certifications and take them while getting paid. I was thinking about getting into web development. MCSD certs are mostly a joke. I have a bunch but mostly because my employer needs me to have them to keep our gold partner status.
|
|
#
?
Sep 24, 2015 04:05
|
|
|
Jsor posted:I think it mostly depends whether you're doing government work or not. I know a few people employed by either the Department of Energy or DOD/Military as software engineers that needed to get certs to be hired, or were hired on a conditional basis that they passed a certain certification exam. Aren't certs more of an IT thing? I know folks that have gotten IT jobs based entirely on having CCNA and nothing else.
|
|
#
?
Sep 24, 2015 05:25
|
|
|
authwiggidywack posted:Going to be inbetween projects at my current full time job, i'm basically been given the opportunity to study for some certifications and take them while getting paid. I was thinking about getting into web development. When I was first asking my boss about moving into technical management, he suggested certifications as a possibility to demonstrate technical ability as a junior developer who isn't interested in the senior dev -> architect type of career trajectory. When I tried to pursue the matter further and ask which certifications in particular he thought would be valuable to the company and worth spending some training budget on, we beat about the bush for a while and then arrived at the conclusion that certifications are pretty close to useless for a generalist. It might be different if you're really into databases, or security, or some other subfield where certifications might be taken more seriously - or if you're at Bigcorp and Bigcorp wants you to get some particular cert for political/regulatory reasons.
|
|
#
?
Sep 24, 2015 05:41
|
|
|
I'm a digital artist trying to break into the world of programming. I just had a phone interview for a job that I think went pretty well and I wanted to get some opinions on what I should ask for pay-wise. My qualifications: I have over 5 years of experience working as a graphic designer and visual effects/motion graphics artist for tv, movies, and software. Over the last year I have taught myself JavaScript and released a full-featured Android game with a responsive single-page design. The job: The job was listed just as "Graphic Designer" but they said that knowing JavaScript and jQuery were a huge plus, as was experience developing web applications from the front end. During the phone interview I found out that the job involves going to meetings with clients and working with them on planning out the designs. It also includes a fair amount of self-directed work and implementing code. I would be creating all the art for every part of a project, from printed marketing materials to websites to apps. This is a full time position with hourly pay. To me this job sounds very close to the responsibilities I see described for Senior Graphic Designer positions with an added dollop of Front End Developer. The descriptions of what they're looking for is basically exactly me and I have all of their nice-to-haves and highly-desireds they listed. I'm not totally sure what the pay for a position like this would be in Portland, OR. I tried looking up salaries but I get wildly different numbers depending on the site and none of the positions they list salaries for really match up to this exactly. One of my friends took down all the information and ran some numbers and told me that I should ask for $40 an hour in the hopes of getting $35 an hour. Does that sound reasonable? I'm open to other suggestions.
|
|
#
?
Sep 24, 2015 20:31
|
|
|
authwiggidywack posted:Going to be inbetween projects at my current full time job, i'm basically been given the opportunity to study for some certifications and take them while getting paid. I was thinking about getting into web development. Cicero posted:In the programming world, certs are generally useless. I've heard of some being useful in specific contexts, and also of a negative bias against them from other employers. Many competent software engineers seem to view them on a resume as a signal that the applicant is clueless. Ithaqua posted:MCSD certs are mostly a joke. I have a bunch but mostly because my employer needs me to have them to keep our gold partner status. Yeah, MS cert exams are especially a joke. It's practically understood that you download the test key ahead of time and just study that and then go ace it. Because even if you have a large degree of practical experience in the technology you're testing for, you'd have to know it inside and out for every obscure thing you may need in order to pass without just studying the test itself. Sure you might learn a handful new things while doing it but it's like a 95% chance you're not learning anything you'll actually end up using aside from answering that question on that exam.
|
|
#
?
Sep 24, 2015 20:52
|
|
|
|
| # ? May 18, 2024 00:29 |
|
|
Tip posted:I'm a digital artist trying to break into the world of programming. I just had a phone interview for a job that I think went pretty well and I wanted to get some opinions on what I should ask for pay-wise. I'm a developer in Portland. I don't know the graphic design world, but in terms of being a reasonably talented and reasonably experienced FE developer (especially one with effects/animation experience) I think $40 would be on the lower side. Consider that it's all the art, all (?) the code, and a lot of customer-facing responsibility.
|
|
#
?
Sep 25, 2015 07:40
|
|