|
one of my co-workers had been stuck on a release for weeks and he's a dev while I'm a senior dev so i feel like i am supposed to be doing something to help but I'm kinda at a loss if he isn't being forthcoming in stand-up, what should i be doing to help? i feel like the poor guy is probably miserable
|
#
?
Feb 23, 2021 16:25
|
|
|
|
| # ? Apr 27, 2024 02:49 |
|
|
what have you tried doing so far? how does he respond to your offers of help?
|
|
#
?
Feb 23, 2021 16:27
|
|
|
Ask if he needs help? That seems like the obvious solution no? Edit: A lot of programmers tend to be embarrassed asking for help. An easy way of breaking that cycle is to say something along the lines of: We are a team and we don’t leave each other to suffer, how can I help? Let’s work through this together and in the future if I need help, I can lean on you.
|
|
#
?
Feb 23, 2021 16:27
|
|
|
You can just directly ask them how they're doing and if they want help. If you aren't their boss they probably will interpret it well and appreciate it. If you don't want to directly ask them, you can say you have spare time to help others in your own standup, then let them ask you.
|
|
#
?
Feb 23, 2021 16:28
|
|
|
ask them how it's going and where they're at with a solution? one of my unhelpful traits is going too deep or focusing on something useful but adjacent to the ticket. if I have to take a step back and explain to someone what I'm doing, it helps me refocus (or ask for help, since someone's right there) and maybe make it clear you're not worried about how long it's taking, as it doesn't sound like you're getting impatient
|
|
#
?
Feb 23, 2021 16:32
|
|
|
Another helpful thing to do is to see if they are overwhelmed? Either the ticket is too encompassing and needs to be broken into a list of smaller tasks, or perhaps there is something going on in their personal life. The worst thing you can do is to remain silent.
|
|
#
?
Feb 23, 2021 16:34
|
|
|
also make sure you're not condescending or assholish about it, and that no one on your team is about it too, people won't ask for help if you're gonna mock them
|
|
#
?
Feb 23, 2021 16:52
|
|
|
Share Bear posted:also make sure you're not condescending or assholish about it, and that no one on your team is about it too, people won't ask for help if you're gonna mock them yeah the key point to this part is not doing it in front of others, ESPECIALLY management/business types. reach out to them privately and ask if you can help with anything
|
|
#
?
Feb 23, 2021 17:21
|
|
|
the main thing that keeps happening is that they tend to solve problems by adding on more stuff to fix the other stuff until the release is a huge messy combination of tickets and it gets delayed over and over my work is pretty much 100% characterized by me suggesting things and people not outright disagreeing (which would be great, we could talk about it!) but also not doing the thing i said at all i've been trying to get everybody to focus on small, regular releases since i started here last march and they just... don't I offered up my help today but the main problem imo is that he is always adding stuff so he is constantly in a state of "almost done" with the new addition so he doesn't think he needs any help
|
|
#
?
Feb 23, 2021 20:17
|
|
|
awesomeolion posted:That's exactly what i'm finding so far! Currently in the investigation phase... goal is to determine the best option for this app. we dont have a big team or any true backend people so i'm weary of building it ourselves... It's not trivial but definitely not as hard as you might think considering the lack of a basic/straightforward library. Android and ios will both give you a wav/LPCM stream from the mic to send to the server. The server just has to sum the incoming streams and send the result to each client for playback. There are some tricky bits around the edges but that's all there is to it at a high level. We already had lobbies/groups and UDP connections to clients set up so didn't have to worry about that part. Detecting silence is probably the hardest part, but important since that's what allows you to stop streaming for a bit and "catch up" buffered/laggy connections without cutting out. You could probably do some fancy waveform/audio analysis for that but I was able to get away with some dumb heuristic.
|
|
#
?
Feb 24, 2021 03:22
|
|
|
cool av posted:It's not trivial but definitely not as hard as you might think considering the lack of a basic/straightforward library. Gotcha, thanks for explaining. One more question for you, did the cost of your custom solution end up being a lower than $0.0017 per person per minute which Amazon Chime offers? That's the cheapest option I can find that looks decent.
|
|
#
?
Feb 24, 2021 04:24
|
|
|
cool av posted:It's not trivial but definitely not as hard as you might think considering the lack of a basic/straightforward library. Wouldn't sending raw uncompressed audio use extreme amounts of data?
|
|
#
?
Feb 24, 2021 04:40
|
|
|
xtal posted:Wouldn't sending raw uncompressed audio use extreme amounts of data?
|
|
#
?
Feb 24, 2021 05:04
|
|
|
thats 88 kbits per channel. not a ton, but you can cut that in half or more with modern wideband codecs that will also sound way better. the trade off being higher processor usage.
|
|
#
?
Feb 24, 2021 05:15
|
|
|
Shaggar posted:thats 88 kbits per channel. 
mystes fucked around with this message at 05:22 on Feb 24, 2021 |
|
#
?
Feb 24, 2021 05:20
|
|
|
g711 is 88 kilobits per sec (64+overhead), not bytes.
|
|
#
?
Feb 24, 2021 05:22
|
|
|
Shaggar posted:g711 is 88 kilobits per sec (64+overhead), not bytes. mystes fucked around with this message at 05:26 on Feb 24, 2021 |
|
#
?
Feb 24, 2021 05:23
|
|
|
oh right. lol. 16bit 44khz. but yeah 88kB/sec would be way too much and you'd absolutely want compression.
|
|
#
?
Feb 24, 2021 05:26
|
|
|
ulaw fuckin sucks but carriers are all lovely so they're never gonna give up their 30 year old hardware
|
|
#
?
Feb 24, 2021 05:27
|
|
|
I mean obviously it's dumb to use uncompressed pcm when you could use something else that would use a zillion times less bandwidth but I just meant that if you're making a toy app you could just use raw pcm and handle a small number of users if you really wanted to as a proof of concept, which it sounds like was what was being discussed.
|
|
#
?
Feb 24, 2021 05:28
|
|
|
yeah you could always swap it out with something compressed later
|
|
#
?
Feb 24, 2021 05:29
|
|
|
i just really loving hate pcm in voip
|
|
#
?
Feb 24, 2021 05:31
|
|
|
doing work for a client, requires me to do all the work in some citrix environment, where i'm allocated very little resources. they use eclipse (lol) and subversion (lol) and they have a ton of anti-malware stuff running, so it's all dogshit slow. went to synchronize my local repository with the subversion upstream, the project was maybe idk, 1000 files. took 2 hours to complete, the eclipse instance was locked the entire time. i can't wait until i'm yanked off this project when i blow up and tell the client its by far the worst possible environment we've ever had to touch in our lives.
|
|
#
?
Feb 24, 2021 17:48
|
|
|
awesomeolion posted:Gotcha, thanks for explaining. One more question for you, did the cost of your custom solution end up being a lower than $0.0017 per person per minute which Amazon Chime offers? That's the cheapest option I can find that looks decent. it's not separated from our other server costs so I can't give a real number, but not enough to notice a difference with a few thousand users who opted to use voice. In any case I don't exactly recommend rolling your own if you've found a library that looks suitably usable. I just checked and I used 16bit mono at 8kbps (small b) compressed with gzip like a simpleton and nobody complained about quality so... idk. you probably wouldn't want to stream music to your friends on it. edit: sorry, 8000 samples per second so I guess it's 16KBps
|
|
#
?
Feb 24, 2021 23:47
|
|
|
is there a more recent version of what's in the art of unix programming that isn't written by that guy i sort of can't take it seriously
|
|
#
?
Feb 25, 2021 16:44
|
|
|
Share Bear posted:is there a more recent version of what's in the art of unix programming that isn't written by that guy What's in that book? I try to avoid esr. I liked the richard stevens books, but they aren't new.
|
|
#
?
Feb 25, 2021 18:14
|
|
|
good news, read() hasn’t changed significantly since Stevens wrote APUE. there’s a lot of Linux specific stuff it doesn’t cover ofc, and I don’t know a good book for them
|
|
#
?
Feb 25, 2021 18:17
|
|
|
leper khan posted:What's in that book? I try to avoid esr. https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s06.html this is essentially the entire book but with supporting details for each part a lot of it is a "it depends" application of stuff, other things are really natural and understandable i find the "supporting details" part to not be particularly enlightening and also we all know the cringey stuff about the guy so its hard to take this kind of voice seriously
|
|
#
?
Feb 25, 2021 18:51
|
|
|
CRIP EATIN BREAD posted:doing work for a client, requires me to do all the work in some citrix environment, where i'm allocated very little resources. They don't sound like people who are going to care all that much.
|
|
#
?
Feb 25, 2021 19:52
|
|
|
hope you get paid by the hour, op
|
|
#
?
Feb 25, 2021 22:48
|
|
|
Shaggar posted:ulaw fuckin sucks but carriers are all lovely so they're never gonna give up their 30 year old hardware yes, all our stuff supports nice modern high bandwidth codecs but what's the point as soon as it hits the PSTN or mobile network it's compressed to poo poo and we're just adding latency for no reason cowboy beepboop fucked around with this message at 10:46 on Feb 26, 2021 |
|
#
?
Feb 26, 2021 10:43
|
|
|
cool av posted:it's not separated from our other server costs so I can't give a real number, but not enough to notice a difference with a few thousand users who opted to use voice. In any case I don't exactly recommend rolling your own if you've found a library that looks suitably usable. well if it works, it works, but 8bit audio at 16KHz will sound vastly better than 16bit at 8KHz
|
|
#
?
Feb 26, 2021 10:45
|
|
|
Dylan16807 posted:well if it works, it works, but 8bit audio at 16KHz will sound vastly better than 16bit at 8KHz iirc some android devices would only give a 16bit stream from the mic, i dont remember why not squish it though...seems pretty obvious maybe I was drunk
|
|
#
?
Feb 27, 2021 02:18
|
|
|
my stepdads beer posted:yes, all our stuff supports nice modern high bandwidth codecs but what's the point as soon as it hits the PSTN or mobile network it's compressed to poo poo and we're just adding latency for no reason atleast w/ teams its all pretty high quality audio until a boomer finds the dial in
|
|
#
?
Feb 27, 2021 04:58
|
|
|
God drat it, my local Boomer coworker frequently calls in while walking back from somewhere and, naturally, doesn't loving mute their phone.
|
|
#
?
Feb 27, 2021 16:29
|
|
|
Shaggar posted:atleast w/ teams its all pretty high quality audio until a boomer finds the dial in
|
|
#
?
Feb 27, 2021 16:40
|
|
|
ctps: I am trying to extract keyframes from video using ffmpeg (actually the C library parts, libav*) and this is hard af.  The documentation is not great and examples were written across the last 20 years, so they do the same thing very differently, as new helpers were added to the interface...I ended up with something that seems to work for some of the videos: I iterate over all the frames by decoding them from packets, and check whether the decoded frame is marked as keyframe. If I count the keyframes, I get the same counts as what I get with ffprobe, so it isn't completely broken I guess. Now comes the problem: it works for two x264 videos I tested first, but for some random x265 video I have lying around, ffprobe thinks that there is 30992 keyframes in the 21:32 video, which would mean that every single frame is a key frame. My own code says 307, which seems kinda low, but potentially plausible. I also tried to build my keyframe extraction through av_seek_frame, which worked for x264 and completely broke for x265, and the API is terrible anyway (do you use offset in pts? dts? apparently that depends on the exact demuxer you are using and isn't actually documented anywhere  )-----edit------ Okay, if I update ffprobe from ancient to less ancient, it also returns 307 key frames. This makes me somewhat happier about my code, but I still have no idea of the whats and hows to av_seek_frame.
Xarn fucked around with this message at 23:38 on Feb 27, 2021 |
|
#
?
Feb 27, 2021 23:28
|
|
|
i’m not sure if this is the best place to ask this question and this is kind of a “help me with my homework question” but regarding content on the web, what is going on that’s “new” as far as presentation, forms of content, etc. the goss on what’s on the frontline words, pictures and videos are standard of course. is there anything tjay state of the art websites are doing to present content in new and engaging ways?
|
|
#
?
Feb 28, 2021 08:18
|
|
|
no
|
|
#
?
Feb 28, 2021 09:17
|
|
|
|
| # ? Apr 27, 2024 02:49 |
|
|
the last big thing i think was where scrolling caused the screen to animate instead of actually scrolling, but that’s a few years old now
|
|
#
?
Feb 28, 2021 10:03
|
|