|
Wheany posted:we use punchcards. well not literally, but a web form where you have to click on a button when you arrive and leave. This is pretty much what I have except instead of a punch-in/punch-out system, I have my HipChat availability status. Luckily yesterday was sweet because my boss's boss was out for the afternoon and the whole support team is offsite visiting another team. I have recently turned up the idle time on my HipChat, though. I have to leave my machine unlocked when I leave, but I now have like a 30 minute window where I can do whatever I want, now.
|
# ? Apr 13, 2016 14:45 |
|
|
# ? May 25, 2024 18:16 |
|
we used to have timesheets where we recorded exactly how long we spent working on each task i know i used rand() to fill mine in and i guess i wasn't alone because management suddenly realised the whole thing was pointless and told them nothing of value and, in a startling display of common sense, dropped the whole idea
|
# ? Apr 13, 2016 15:03 |
|
i said "gently caress that hdl poo poo" and went back to playing with haskell and hdl parsing and i successfully parsed out all of the register/constant declarations from one of my files, including things like ranges defined by expressions this was very exciting for me
|
# ? Apr 13, 2016 15:14 |
|
so I found an RPC server plugin for kerbal space program brb going to turn my Arduino (which I bought a new CPU for after I toasted the last one) into a dedicated KSP autopilot Surely if they did it in 1969 I can do it in 2016
|
# ? Apr 13, 2016 15:16 |
|
Luigi Thirty posted:so I found an RPC server for kerbal space program we have at least a dozen employees still here from when we did it in '69
|
# ? Apr 13, 2016 15:17 |
|
I have to track how much dev time I spend on work items and I hate it because no matter how much management says they're not making judgements based on it, I know they are, and I've got too much baggage from past jobs of getting the classic "Why did you spend 40 hours on this task? It really shouldn't have taken that long." Thanks for your input, manager who hasn't touched the code or have any idea how loving hard something was to do. gently caress time tracking
|
# ? Apr 13, 2016 15:18 |
|
Finster Dexter posted:no matter how much management says they're not making judgements based on it wait, what are they claiming is even the point then
|
# ? Apr 13, 2016 15:42 |
|
Soricidus posted:wait, what are they claiming is even the point then at my place -- for improving time estimation statistics
|
# ? Apr 13, 2016 15:44 |
|
Bloody posted:we have at least a dozen employees still here from when we did it in '69 neat
|
# ? Apr 13, 2016 15:45 |
|
megaparsec seems to let you capture the parser state as its doing stuff. i wanna do that so i can tag my parsed bits and pieces with where in the source file they came from so i can give nice descriptive messages after parsing an entire file but their tutorials do not cover it and i am dumb. looks like there's a runParser' that somehow exposes the state as it goes along? i have no idea what i am doing lmao and every google search containing "megaparsec" gives a bunch of space related bullshit
|
# ? Apr 13, 2016 15:54 |
|
hackbunny posted:oh, lucky you, our OS class was all about fork and sysv ipc, the project was about writing a silly toy task dispatcher. in our study group we called it bogotron 2000 because it did a lot of work to accomplish nothing . I had the honor of having to explain the name to the professor lol is ipc inter-process communication? i didnt have to do that, but we did implement fork as well. this is everything i did for that course: *implement fork and exec *implement a bunch of other syscalls that escape me *implement a ticket-based process scheduler *implement a basic shell (really basic, really really basic, i wasn't allowed to use a parser library ) *implement a slab allocator *change the virtual memory layout (the kernel we were using put the stack in the middle by default, we moved it to the top of the memory range and put the heap in the middle) *implement spinlocks and semaphores *implement an fsck tool for a super basic filesystem *implement a few basic programs like cat and grep against a different filesystem *implement multithreading i was working full time throughout all my cs courses too, since epic was paying for them. i think i got like 2 hours of sleep a day max during that os course. Soricidus posted:we used to have timesheets where we recorded exactly how long we spent working on each task we technically have this time log to fill out, but my boss doesnt give a poo poo as long as theres at least 40 hours in there logged to "development" and i remember to log the billable poo poo accurately. thats really variable though and some people have hideous micromanagers that want 15-minute level detail of all the poo poo you do.
|
# ? Apr 13, 2016 15:56 |
|
Soricidus posted:wait, what are they claiming is even the point then Barnyard Protein posted:at my place -- for improving time estimation statistics Yup, exactly this. Time estimation is easily my least favorite thing about this line of work.
|
# ? Apr 13, 2016 15:58 |
|
i ended up spending all day yesterday getting one dialog to work and i got five tickets back about it still being broken as gently caress just now if you asked me for an estimate before i started i would have said 2-3 hours tops
|
# ? Apr 13, 2016 17:17 |
|
it usually takes a few hours for me to come up with an estimate the right way. i'm anticipating being asked to give an estimate for the estimate.
|
# ? Apr 13, 2016 17:25 |
|
HoboMan posted:i ended up spending all day yesterday getting one dialog to work and i got five tickets back about it still being broken as gently caress just now always triple your estimates
|
# ? Apr 13, 2016 17:29 |
|
LeftistMuslimObama posted:is ipc inter-process communication? yes, sysv ipc in particular is an ancient and especially lovely ipc api LeftistMuslimObama posted:i didnt have to do that, but we did implement fork as well. this is everything i did for that course: I'm envious! all we did was write the aforementioned bogotron 2000 and run it on the solaris machines of the unix lab. I wish we had toy kernels to play with, I had to learn all that stuff the hard way on my own by working on reactos
|
# ? Apr 13, 2016 17:34 |
|
I'm looking for a fully managed PaaS that I can just ship code to, and I'm willing to spend the money. Is heroku still the best option for this or is there some new stuff from azure or openshift worth looking at? e: python 3.4 or 3.5, postgis, background celery procs
|
# ? Apr 13, 2016 17:54 |
|
use azure and asp.net.
|
# ? Apr 13, 2016 18:02 |
|
Barnyard Protein posted:it usually takes a few hours for me to come up with an estimate the right way. i'm anticipating being asked to give an estimate for the estimate. I didn't know George costanza had a forums account
|
# ? Apr 13, 2016 18:07 |
|
mostly I just want to torture myself by writing a cooperative multitasking kernel on my arduino because I never got to take any operating system classes in college
|
# ? Apr 13, 2016 18:10 |
|
Luigi Thirty posted:mostly I just want to torture myself by writing a cooperative multitasking kernel on my arduino because I never got to take any operating system classes in college I did a preemptive kernel for dsPIC a while back for the same reason -- its really fun and surprisingly simple to get something barebones basic working, although i learned that the simple-barebones design didn't lend itself to the kinds of constructs you'd want in a real kernel. i read this book a while back, its about coding up cooperative statemachines, but it goes into a lot of OS ideas suitable for a small platform like an AVR or PIC -- he starts selling his licensed framework pretty early on, but he describes the fundamental concepts well enough that you can design and write your own http://www.amazon.com/Practical-UML-Statecharts-Event-Driven-Programming/dp/0750687061
|
# ? Apr 13, 2016 18:41 |
|
If it needs structural changes to the db, rest interface and ui, the default estimate is 2 weeks.
|
# ? Apr 13, 2016 18:48 |
|
#noestimates
|
# ? Apr 13, 2016 18:49 |
|
Barnyard Protein posted:it usually takes a few hours for me to come up with an estimate the right way. i'm anticipating being asked to give an estimate for the estimate. we call that a scoping exercise. sometimes we spend more time coming up with estimates than actually producing anything.
|
# ? Apr 13, 2016 18:50 |
|
Wheany posted:If it needs structural changes to the db, rest interface and ui, the default estimate is 2 weeks. I wish those were separate concerns in this code base. A lot of ASP.NET web forms pages with codebehind, and the codebehind has everything, db access, html manipulation, 3rd-party API requests. The best is when I work on poo poo that this one guy did with his ninth level of hell jarvascript implementation. Then it's like all the normal poop wrapped in a thick layer of diarhea sauce
|
# ? Apr 13, 2016 18:53 |
|
web forms is so gross. can you imagine that there was a time when it was the best thing out there?
|
# ? Apr 13, 2016 18:54 |
|
well i found the page where everything just throws an exception instead of trying to do anything graceful this is literally the user entry validation: C# code:
|
# ? Apr 13, 2016 19:13 |
|
Shaggar posted:web forms is so gross. can you imagine that there was a time when it was the best thing out there? as someone who is working on nothing but gross webforms, no i can feel myself forgetting what MVC is a little more every day
|
# ? Apr 13, 2016 19:14 |
|
HoboMan posted:as someone who is working on nothing but gross webforms, no loving same fml
|
# ? Apr 13, 2016 19:15 |
|
Shaggar posted:web forms is so gross. can you imagine that there was a time when it was the best thing out there? when it was a thing you were better off with php than webforms
|
# ? Apr 13, 2016 19:19 |
|
when i first used webforms after using php i was amazed by how much better of a model it was. this lasted about a week before i concluded that the webform model was actually so broken that even php soup which mixed sql and html was better
|
# ? Apr 13, 2016 19:50 |
|
akadajet posted:when it was a thing you were better off with php than webforms lmao no.
|
# ? Apr 13, 2016 19:51 |
|
https://www.codingame.com/start i'm trying my hand at this but i only know HTML and XML, the 2 programming languages that I've heard of
|
# ? Apr 13, 2016 20:08 |
|
Beast of Bourbon posted:https://www.codingame.com/start lmao at implying HTML/XML is a programming language
|
# ? Apr 13, 2016 20:13 |
|
more like dICK posted:I'm looking for a fully managed PaaS that I can just ship code to, and I'm willing to spend the money. Is heroku still the best option for this or is there some new stuff from azure or openshift worth looking at? heroku or go into the aws ecosystem. elastic beanstalk for your app + RDS for db + apparently you can use SQS with celery. your stack as specified wont let you benefit from most of azures PaaS features so youd end up mostly using it as an IaaS provider and janitoring the vms yourself.
|
# ? Apr 13, 2016 20:18 |
|
HoboMan posted:lmao at implying HTML/XML is a programming language the L stands for language, C++ doesn't have an L in it, so you could make the argument that it's not a programming language. although maybe it's like college majors and if you have to put 'science' in the title, then it's not a science. like political science. hmmm
|
# ? Apr 13, 2016 20:18 |
|
HoboMan posted:lmao at implying HTML/XML is a programming language please don't be that guy
|
# ? Apr 13, 2016 20:20 |
|
blup blup Programming Must Be Difficult blup blup
|
# ? Apr 13, 2016 20:22 |
|
honest question: is html turing-complete
|
# ? Apr 13, 2016 20:22 |
|
|
# ? May 25, 2024 18:16 |
|
Luigi Thirty posted:i'm a 10x programmer im a log 10x programmer
|
# ? Apr 13, 2016 20:24 |