JB Transcript Accuracy comparison

00 JB Transcript Accuracy comparison.md

I'll use this gist to compare a few methods for transcribing coder radio 428 with multiple engines

01: A combination of Youtube for transcriptions and IBM Watson for speaker diarisation

02: AssemblyAI

01_YT_Watson.md

Chris: this is coda radio episode 428 for august 23rd 2021 [Music] hello friend and welcome into jupiter broadcasting's weekly talk show taking a pragmatic look at the art and the business and the whole world of technology around software development this episode is brought to you by cloud group you know cloud guru has the cloud playground that means azure aws and google cloud sandboxes on acg's credit card not yours so go get certified get hire get learning at a cloud guru.com my name is chris and joining me like the podcast pro that he is it's our host mr dominic hello mike

Mike: hello mr fisher

Chris: podcast time yes here we are i am uh i am excited to be in colorado springs we went eight days off-grid no hookups running off solar and so now it feels luxurious to be in a campground parked into parked and plugged into power and like have air conditioning and stuff it's like i'm fancy all of a sudden however it's like a downtown city here this this campground is so freaking crazy busy so apologies if i've got weird sounds in the background because there's a whole party going on around here are you fired up about the podcast though you ready to go

Mike: i've been looking forward to this i've been waiting for this one a long time yeah well i mean i i kind of been

Chris: the opposite i kind of been dreading it because the inbox this week the inbox mike it was so bad we got so many emails but good it was a good problem to have and so we thought we should definitely try to get to some of these and uh listener chris not me but listener chris was fired up about crystal he says hey guys i'm a little bit behind but i just listened to the ruby and the rough episode and i have to plug crystal at crystal lang.org they released version one not that long ago it's written like ruby but compiled down and fast like c i've started to rewrite some of my ruby in crystal and it's been a huge success thanks for what you do and sorry i haven't chilled crystal before i'm just now catching up okay so i really uh i i really didn't know what else to say other than i guess thanks listen to chris for firing firing an email about crystal it was totally off topic but felt like we should include it at least well it's actually

Mike: not that off topic if you remember the dark days of the coding challenges it was one of the languages that wes and i used to torture each other ah yes

Chris: right

Mike: if i had done my homework i would have been able to find the episode but unfortunately uh we're fortunately there's just a lot of quota radio out there in the universe and i could not find it but it's there somewhere we need some

Chris: sort of system that goes through and indexes all of the speech you know like a does a transcription and then generates keywords why is there not like

Mike: a google search of podcast content they

Chris: were playing around with it for a while and then they just sort of dropped it

Mike: really yeah because you know there was a i wanted to reference something for today's show from atp uh something john cercus said and after like an hour and a half of trying to figure out where he said it in which episode and thank god they used chapter markers i still could not find it because it turns out i was a year off in my looking and i was like i tell with this i'm not i'm not doing this

Chris: yeah there i i there probably are a few other services out there there's a there's companies out there that are doing it to monitor when brands are mentioned because it's a smaller problem scope right they can just monitor for netflix sure makes sense so there's businesses around that so i think it'll expand from there all right well jim wants to know more about c plus plus deets from you he says you know last week mike tossed out one of his classic throwaway saying that he's running c plus on his back end and well are we just supposed to ignore this please share as much as you're able what mode of running what libraries what inner motivation led you to make this choice are you running c plus and cgi and websockets dark matter c plus plus developers are slowly steering their organizations away from cute native apps and i want to know what you're doing yeah so so definitely

Mike: not cute right this is the server-side stuff i mean very simple very kind of pure ansi c plus here just cmake uh the id some of us are using c-line i'm using vs code because i've just drank that kool-aid like it was going out of style the way we're using it is maybe less web servery than either i made it sound or or that you got the impression of it's more we have python so the stack is like okay we have an application most of the interaction you know the webby's stuff isn't flask for the most part and anything performance is in c plus it was going to be rust

Chris: interestingly enough but finding

Mike: external help for rust is way too hard and way too expensive right now so we kind of fell back to c plus plus i have to say we are we have taken a i think a little bit of a weird approach so one because we control everything this is our own thing it's the newest standard c c plus 20. so we're using all the fun little new bits right the uh the smart pointers all that good stuff we're also not using boost which i think is important to mention we're kind of doing things just as they are standard library everything else is kind of written ourselves i know that smacks of not invented here but there are reasons that i think probably over the next six months we'll talk about but why for something like this where we want to control it uh full stack and it really is full stack meaning from server to hardware to not um to not have that level of dependency so that kind of segues into your other point dependency management uh other than like the standard stuff you would have in c make we really don't have many dependencies that may not be true forever and you know i know i know there's options but

Chris: this is kind of where we need

Mike: performance and don't get me wrong where we can we're doing everything in python but sometimes that's not applicable i know there's a ways to compile python down i look at them i get it it's just a little too voodooy for me easier to write a linux c plus binary and huzzah call into it right sure

Chris: well thank you for sharing that uh our next email is asking for us to consider intel for a moment

Mike: i didn't bring my tissues

Chris: i know yeah i've added your beer to pour one out i need a 40. jacob writes can you guys talk about intel for a few minutes i'm with chris that konom is deploying solutions that widen the chasm with their audience but if it was a contest i would think intel would be able to take the cake i've never seen a company deploy more promises than actual products while they will likely always stay relevant for a large audience how do you see their future unfolding do you think apple's upcoming m1x or m2 processors will encourage a double-digit shift from the x86 architecture our thoughts on intel mr dominic what do you think about intel's future and their products like i was thinking about our friends over at system 76 they tell me they're working on a laptop that they want to have out in a couple of years right and i thought to myself it could be really awesome you know metal chassis you know maybe it's milled from a block of aluminum runs linux has replaceable parts and then i thought oh but it's probably going to be an intel chip or at least an x86 chip and that may be a deal breaker for me because it kind of feels like buying a gas guzzler in an era of electric cars

Mike: i mean what's the time frame here is it like now or is it five to ten years i mean i

Chris: think starting now but the way jacob defined it it sounds like with the m1x or the m2 assuming there's some good performance gains there there i think

Mike: it's uh and i know we get a lot of heat for this but there's no way to i think you know objectively look at the situation and not say that the m1 the class of processors from apple isn't a leap forward which you know necessarily means that intel needs to catch up i know my broader thoughts are intel made a bet like five ten years ago that didn't pan out right and they got kind of hosed they're not going to go away they're a big company they'll be fine can they catch up sure right they could have the next good idea for processor technology is it going to happen within the next let's say two three years i don't think so but keep in mind apple's not gonna like license their processors out now you might be thinking okay a risk architecture um like you know those type of processors and maybe iot that's true that's a market that i don't think intel competes in very well right you're with your iot devices your kind of automation factory kind of stuff you're probably running arm and i know um uh scif sofi sci-fi the company chris slatner went to from tesla chris liner for those who don't know is the guy who made swift at apple um that's what they do right they do like bespoke um risk five processors for iot devices that's also a competitor i mean on the desktop i i sort of feel like that's the market we all all us nerds care about quite a bit but kind of doesn't matter

Chris: i don't know

Mike: also those ryzen chips are looking pretty good and uh that's the wild card

Chris: here right is maybe amd pulls it out maybe intel with their arc gpu and just really gets off their ass and actually manages to produce something but i suspect with intel it's a corporate structural issue i think the rod is from within um and they just they got complacent again and their management structure changed and now they cannot fix their problems at least that's what i worry about and i have to say i'm not sure yet jacob you really i think i'm on board with what mike says is i need to see where ryzen plays out i need to see what some of these with these amd laptops with amd gpus in them what they look like because we've seen a few on the market but we haven't seen it at the scale that we've seen the intel stuff yet i did note when i was talking to the system 76 crew my win my thought was oh but it it's still gonna be intel right so it's almost like doesn't matter uh and i hadn't like conclusively reached that but it it started it's starting to feel that way and i'd be curious to know if the audience out there feels similarly like it's one thing i think too when you're thinking about it from like a theoretical standpoint like oh yeah if i'm gonna go blow two grand on a laptop you can just sit there and you can throw around what you may or may not do all the time it's easy but when you at what it actually comes down to i can only buy a laptop like every four or five plus years like it's not something i do all the time i'm this is like my one chance to get it right i want something with a good screen i want something with a good build quality i want something with a good keyboard i want something with good battery life and i want something that has high performance and it's my actual money now that i'm spending i find it getting harder and harder to buy an x86 based system thankfully i'm just in a wait and see mode right now c sharp hater writes in though he says anyways i can't help but take those stack overflow surveys that you guys cover as just a chance to rag on c sharp even though it's not my favorite language objective c is a bit overheated in my opinion c sharp on the other hand is an evil virus from hell forced upon the human species i've never encountered a language that was far more removed from what it claims to be in the case of c sharp it's quote easy to read and easy to use but that phrase couldn't be a better description for everything c sharp is not the nastiest and most unpredictably dysfunctional code i've ever seen in my life has always been written in c sharp it's ugly it's easily breakable and you could never rely on most people to write anything maintainable the best c-sharp can do is work in a clunky way i was shocked when i found out that c-sharp was sixty percent loved but the more i thought about it i thought about is probably a blessing really that many programmers seem to either not remember how terrible c sharp is or because they've long abandoned it for the browser i never forget the nightmare that is c sharp especially back before stuff like vs code existed and i had to use the old visual studio institutionally forced upon me there may be worse languages objectively but c sharp will always have a special place in my heart like as much as i hate javascript for spawning electron at least electrons cross platform and is actually easy to use in fact many of its problems stem from being so easy to use that a lot of people can build apps that just work well enough and never had to learn good programming habits c-sharp is the worst i mean i don't know about that uh i think there's some people out there that would strongly disagree i know some people who desperately are in love with c been trying to figure out how to balance their real relationship with their relationship with c sharp so

Mike: yeah i mean so so i i feel for the writer here right because he's got like my level of love for objective c a language that i always say the vast majority of people hate he hates the language the vast majority of people love but the funny thing that he mentions there that he was forced to use visual studio i'd care to wager one of the main reasons lots of people like c sharp is because of visual studio i don't know man i mean it's not really my back right i tend to run mac or pop but if you want an ide that does lots of stuff for you and helps you out granted i know it adds complexity with all those weird config files but i i find any argument this is a visual studio does not succeed in its primary mission of being a great ide for net development

Chris: a little

Mike: nonsensey um and i know that's not the argument he's necessarily making but see a lot of this stuff is about like taste and preferences right like why do i like objective c because i i'll turn this into a positive argument not to bash one c sharp because i actually i actually kind of like c sharp objective c until like the very end and i know end is like in quotes because it's not dead yet i will assert that forever just didn't have that many features like i'm doing a bunch of python now and there's like a million magical ways to do things in python right in a because of like you know different libraries whatever the objective c is like you can send a message to any object we have built in these very few features um then you can import coco fine you get some stuff there but it's a simple language with simple concepts until you crack open the c part of it and then you you know you can it's still simple but you can go cry because you have to do a lot of memory stuff i have a feeling that some people like c sharp because it's a big language with lots of features and just a ridiculous amount of abstraction right where objective c is like hey man you want to do that no problem you can cobble together traditional c blocks and like whatever weird ns auto release pools and object sending things we've got going on here do some dynamic stuff in the run time c sharp is like oh dude we have a class for that yeah just throwing away and just throwing async away and we'll figure it out don't worry about it like that i don't know it's kind of like easy mode right

Chris: yeah and that's not a bad thing it

Mike: should be easy mode okay can i [ __ ] a little bit do it doing tons of python loving it except for both pycharm pro nvs code for whatever reason cannot do uh intellisense on the sql alchemy uh like classes right so if you have a model that inherits from things like db.model and i think we should just taking a step back sql alchemy is effectively active record or entity framework from you know it's a orm as a guy who has a serious typo problem as chris can tell you that has been killing me because i just not being able to hit like dot escape or i think i think i think i have a hot to like dot tab and seeing the options of methods and properties is i spent like three hours this weekend trying to figure out why why it's not working and there's just a bunch of like annoyed people on stacker floyd like i understand why this doesn't work right that's the kind of like quality of life thing that's annoying and if you are a pythonista of of more of a beard than me and you can explain to me why vs code cannot figure when i do like mymodel.query.you know whatever why the dot doesn't bring anything up i will be very happy i think they prefer to be

Chris: called python hounds

Mike: i thought they were i thought they're like snake charmers

Chris: yeah that's right that's right another topic that we covered on the show that uh got some feedback and we're still we're still the door still open on this one is uh transitioning into a manager from a developer to a developer manager and jacob wrote in and said there's one thing that i really recommend is sit down with each person and do a one-on-one session and ask them this question tell me about a mentor that impacted your life he said you'd be surprised at the kind of answers you get and insights you get into what uh motivates people and then peter wrote in with two book recommendations the unicorn project and the effective manager we'll have those linked in the show notes he says they give you actual advice and not just theory and that the unicorn project uh gives you a good idea of like how to run a devops shop but in story form so that's the unicorn project and the effective manager we'll have those linked in the show notes if you got some some of your own snakes that you have to charm um you know those are very useful we have more uh we have more feedback to get to but if anything we've talked about in the show strikes and interest of of yours and you want us to talk about it more or expand on something or you have thoughts or insights go to coder.show contact and share those with us i'm in colorado springs like i mentioned we just did our denver meetup a couple of days ago and there was an incident that i got to talk to you about because there's been some demands that have been made that i need to relay to

Mike: you

Chris: and uh we got to come to some sort of decision so there was a bit of a coder upset at the denver meetup i swear to

Mike: god if you say closure nope nope uh a

Chris: coder quorum was assembled they came up with a list of demands and then they claimed one of the bars as

Mike: their territory so uh that got a little weird

Chris: and um then they wouldn't let any of us linux nerds enter the bar land until they had your blessing and uh they refused to listen and i told them you weren't coming because we assumed you were gonna be moving and they just weren't having it so now they're holding several macbooks um from the participants hostage threatening to spill water on them unless uh you know you show up at a future meetup i don't know how long it's gonna go on i think they're even threatening a hunger strike it's pretty bad the one in denver is the one i

Mike: should have gone to with and by the way just uh for carl um i've gained 400 pounds so don't mind the really big jacket coming into the system 76 factory just it's cool you know i've been eating twinkies

Chris: oh man can i just say that the hardest thing about this meetup is the fact that i gained a bunch of weight during the lockdown stuff and so now i have to go around like i don't fit into my clothes like it's just the whole thing's embarrassing it's just you know i i have

Mike: a bunch of the jb shirts that i no longer fit into like the old coder ones i'm like those are my aspirational

Chris: shirts now yes yes the coder like 100 shirt is like i like

Mike: look at it sadly all the time oh man i

Chris: know i know and you know uh you know what you honestly where it really kicked off for me was uh a year ago we were just going independent like this week because of the week we went independent like tomorrow was the day we announced it oh

Mike: happy anniversary thank you

Chris: and the the months many months that led up to that the negotiation process was extremely stressful and uh it was right at the beginning of the lockdown and all of that and i just put on a ton of weight and i realized oh stress really does make a difference on that kind of thing like i have to i'm gonna have to just take steps in the future when i'm under a lot of stress to kind of like preventative care kind of thing but man oh man was that a process and now here we are here we are doing the meet up and all that kind of stuff it's been great so almost one year to the day as we record by the time this episode comes out it'll have been the year anniversary it's pretty great

Mike: so for folks who might want to plan to go to a future meetup wink wink is there somewhere these are posted or

Chris: i think we're gonna come up with a meetup.com replacement right now we're still using meetup.com okay but i feel like it's time to find something else meetup is is designed around the idea that you're holding meetups in your area it's not designed around the idea that you're holding meetups somewhere far away from you oh by the way optimus grey points out in the chat room that you can go to notes.jupiterbroadcasting.com coder dash radio or just go to notes.jupiterbroadcasting.com and you could search there for our coverage of crystal there's a really nice index of all of our previous shows with a special search front end in fact the coder audience might find actually pretty clever so that's notes.jupiter broadcasting then it's like a it's quite the fancy search lino.com coder go there to get 100 on a new account and you go there to support this show the year of independent coda radios has been made possible by lynn oden we enthusiastically endorse them they started in 2003 as one of the very first companies in cloud computing 18 years later they've just that entire time focused on that mission they haven't gone crazy they didn't get like some weirdo vc funder in there that wanted them to get into like some crazy aspect of technology they just hunkered down and built something incredible and as the different things in the market evolved lynnode has consistently responded and over delivered and that's the part that i have been fortunate to witness over the last couple of years like i'll tell you something they've added kind of more recently that when i first started using the note i never thought they'd have and it's bare metal boxes i just never thought that was going to be their game but they looked at it you know something customers want and it's something we can do right on top of that though they just have an easy to use powerful cloud dashboard they have 11 data centers around the world you can deploy to so if you need your application to be performant if you need it accessible they've got you covered there they've got ddos protection they've got this beautiful beautiful beautifully fast network because they are their own isp and the reason why i want to mention this is i have set up applications that i have deployed in different data centers that come back to their new jersey data center for object storage or something like that and i was worried that maybe that wouldn't be fast enough but it has been exceptionally well performed the entire time we were on this trip as the crew is recording clips both video and audio which man man were people just like like brent was working his tail off it we would get together in the evening at the airbnb and we would upload all of our various footage to lenode's object storage so we have one centralized place we don't need to bother with any crazy front-end software you just upload it to their object storage anything that can talk s3 can talk to linux object storage they have also entire one-click application deployments of the app of the entire stack if that's what you like or you can do it yourself and their pricing is thirty to fifty percent cheaper than other major cloud providers but with that one hundred dollar credit you're going to get when you go to leno.com coder you can really kick the tires you could try out one of their high-end amd epic dedicated cpu systems that just kicks the butt of everybody out there or you could go for their five dollar a month box and see how far you can push it you'd be really surprised by that and lenode made our jupiter colony reunion road trip possible they they were there at the denver meetup giving away prizes and some pretty special events and memories were created because of their support and now you can get a hundred dollars to go try it out for yourself just go to leno.com coder get 100 on your account and support the show lenode.com coder uh i was a little fired up last week and that's what got the most feedback into the show i was having a real struggles with ns url session d on the mac so quick recap because i i'm a glutton for pain

Mike: i thought to myself

Chris: i don't really have a great computer for this road trip uh i kind of went all in on my x1 carbon at the beginning of the year and frankly i'm just not very happy with the overall performance and battery life other aspects of the machine i am quite happy with i love the build quality i think the keyboard is great it runs linux because i got the fedora edition like a freaking champ like if you needed a mobile workstation to do some web development to get terminal access to check your email to browse the web this thing is checking the boxes all day long

Mike: however

Chris: if you are trying to do a crazy remote live stream broadcast with a dozen different applications including a full digital audio workstation including obs for broadcasting including soundboard software edge browser chrome browser firefox browser all running simultaneously jack and pipe wire running the background routing options doing real-time effects processing and then of course recording each individual device and track to a flac file in real time to the hard drive like that's the baseline before i'm even like browsing the web reading the news stories responding to the chat room or launching slack like that's the baseline is a lot of work for a machine and i i just opted i just said i got to bring the macbook pro because that's the fastest machine i've got

Mike: okay so far i'm with you

Chris: okay so uh i tried it and i thought how bad could it be and i really i really got nailed by something called ns url session damen last time and at the time because i didn't really know what it was it was i was in problem solving mode it was consuming all of my available bandwidth it was just really going to town first it consumed all my downstream and then it consumed all my upstream and it happened for a period of like three days it was really getting to like by the time i've done coder i was three days into this being a problem what i learned after the show is that nsurl session daemon is actually kind of a good idea in principle i think it's actually the implementation that is the problem it's a mac os system process that is used for downloading and uploading data for any application initially i thought it was like just an icloud thing or something just for mac os but now i understand it's actually a class that is made available for any application and any application can delegate transfers to that daemon and the nice thing is it that it will continue even after you quit the application so you can be like telegram and say hey i need to download this thing nsurl session will do it for you the main telegram process is left free you can quit it the download continues and through my investigation basically by throwing lsof and looking at everything it's doing because activity monitor gave me information but by using traditional unix tools i was able to determine like edge uses this telegram uses this of course all the icloud stuff uses this like tons of applications are using this damn ns url session in the background on my mac and that is why it was just constantly doing [ __ ] because i had limited connectivity and so things just sort of queued up and it was just dutifully going through the list and executing its list of things to download consuming all of my bandwidth and we got a lot of emails in that people said yeah this drives me crazy uh hans wrote in said he switched to plasma he switched to kde neon because he was just sick and tired of all the stuff that the mac does in the background i'm just curious to see if you agree like i don't think you've really suffered from this because you probably are somewhere either with high bandwidth or something like you've never been plagued by this problem i don't think

Mike: i've never had this problem

Chris: it could happen to anyone it could happen to any mac user you just kind of need to be like in a low bandwidth situation for an extended period of time and um there were a lot of options that were that came in nobody really wrote in with like a way to use launch control to just disable this thing which is what i actually expected because that was the route west and i took was like well let's use the system application launcher to just shut this thing down but no no what people suggested i do is use a hammer and i ended up doing just that so i checked out two applications that were written into the show one is called lulu by a company called objective c s-e-e yes i know you like that and this will have a link in the show notes is a really neat open source firewall it it watches for outbound connections and then prompts the user if they want to allow or block that connection and you can say temporarily block it or block it forever and it runs at the network level using the newer apis in mac os to make sure that only the software you allow goes out and the ui is fine it's not phenomenal but it's fine and it absolutely gets the job done it's at objective-se.com and it's lulu the only problem i had with lulu is a couple of applications didn't work like transmit and byward and a few other applications just weren't working and i knew about little snitch which is like the og firewall i legitimately may have used little snitch back in like the mac os nine day or eight day like i mean a really long time ago i'd used little snitch so i thought let's go see how they've done now that company's name is objective development so you've got objective c s-e-e and objective development um excellent

Mike: company names both i

Chris: know um and little snitch is some big

Mike: but like the good kind wow i know i know it's it's phenomenal i

Chris: mean just as far as like as far as like an application design it gives you like a real-time norad missile map of like where all your connections are going a beautiful way to see which ones you've allowed or disapproved it gives you prompts that it's aware of certain third-party applications so like if you are another independent software developer and you deny that applications connection to the internet for some of them like say bartender it'll come up with a little disclaimer says hey bro just so you know you know these are good guys uh this is just the auto update you could just go into the preferences and turn that off and still allow this thing access to the network if you want like they've vetted certain independent third-party mac apps and give you additional context when you go to deny them access which is kind of nice they've also made it really simple to understand what stuff is icloud stuff and what stuff is mac os stuff and i just use this thing like full stop to just shut down as much possible on the mac while i'm recording and at this point in time i've nearly blocked 60 different processes from connecting to the internet it is mind-boggling how much crap on the mac connects to apple i mean you click the safari url bar four different background processes connect to apple you launch spotlight three different processes connect to apple constantly all the time i mean the the list is like i said it's 60 different processes long now that come up when i like either launch an application or boot mac os that all phone in to apple and i it reminds me when the linux community crapped their pants over the canonical dash search and how it sent a ping to canonical when you would search for amazon items because literally every move i make on mac os some process is connecting to apple or some just automated sync process is connecting and it's it is really eye-opening to see how much this mac is talking to the mother ship whenever you're doing anything and so i've just blocked all that crap while i'm recording i've just i've created a profile that's like a hardcore profile that tons of stuff is shut down hard and now it's working great i don't know if it's going to cause any long-term damage to my mac but in the meantime i can at least use my internet it's not your

Mike: internet or your computer you need to accept that apparently

Chris: it feels like half-assed on apple's part here because they've created this ns url session daemon that kind of seems like a good idea but wouldn't the next obvious step be to meter that in some way to not abuse a connection you know like isn't that just like the next obvious thing

Mike: so it's server based software and it's apple true

Chris: i mean when you look at all the stuff that just connects to the internet anyways it's clear they give zero craps about the amount of network activity and network connectivity they use they don't

Mike: give it they they have no concept of like i'm using this on a tethered connection now granted there is an api now to tell that you're on a tethered connection and responsible developers could like you know act accordingly right so were you able to ascertain like what exactly they're doing or it's just telemetry right

Chris: in most cases it looks like it's it's checking to see if something is on a deny list it's checking to see if something is signed or it's checking to see if something needs to sync or like with spotlight it's pulling in market information for stock searches and all this crap that i just don't need it's it's preemptively caching a lot of that stuff in the background um or there's a lot of assistant services and siri things that just try to like connect and and be preemptively ready when you click a url bar all that stuff though generates metadata even if it's not intentional tracking it is metadata that gets triggered and if even if it's not apple that logs it your isp could be logging it so i just turned all that crap off yeah that's not good i

Mike: don't even know you know it's just not good and it's kind of a garbage

Chris: experience going through it because for like the first two days you just live with tons of prompts about approving or denying software connectivity and you just have to wait because you can't really proactively go through and do it you kind of just wait until something prompts and then you tell it no and so for like two days now i'm just sitting here saying allow or deny and i think i've caught most of it at this point but it takes a while to get it set right and this whole thing could just be avoided if i could just go under like the wi-fi menu and say low bandwidth mode and then

Mike: you know problem solved because they've

Chris: already done the hard work of consolidating a lot of the background process and downloads into nsurl session

Mike: so it's like just just give me the damn

Chris: option to just throttle that thing just it could just be a little check box in the wi-fi drop-down that says limited connectivity or something for goodness sake there will be no check boxes and

Mike: drop down menus on vacuum no you can't

Chris: have that maybe at best i could get something if i held down the option key anyways i really like little snitch i paid the 47 bucks even though after this road trip i don't know if i'm gonna ever need it again but i figured you know what great software really well designed it's been around forever and now i actually feel like this mac is under my control and it shifts my feeling about mac os like i'm i'm happier with it again it's doing the job for me once again and as long as i just use these tools to brute force it to behave the way i want

Mike: it's fine

Chris: oh man at least it's really cheap hardware and uh doesn't have a lot of ongoing costs so you know datadog.com coder radio go there and sign up and get a free trial and if you create one dashboard you'll also get a free t-shirt yes this episode of coda radio is sponsored by datadog you know they are the monitoring and security platform for developers security and ops teams in the cloud age you can truly unify your metrics your traces and your logs all in one place that means you can troubleshoot issues faster and break down silos between teams i was just visiting with a shop here in denver recently that uses datadog and they had very positive things to say you can create real-time dashboards for over 450 different business technologies in just minutes you can easily pivot from like a high-level overview of your environment zoom in you know zoom in enhance down get granular and visualize the metrics of specific events datadog offers infrastructure monitoring security monitoring user monitoring and it all brings it into one place and makes it easy for you to communicate display and just review so go get started with a free trial today at datadog.com coda radio go see why thousands of companies trust datadog as their monitoring solution go see why so many coda radio listeners are using datadog it's datadog.com coda radio if you start the trial and create one dashboard datadog is going to send you a free t-shirt who doesn't love free swag datadog.com coda radio [Music] so there's a meta topic you and i have been discussing for a while

Mike: and it's it's

Chris: platform control and we've come at this in several different ways we've talked about it in terms of app stores before i think you could even say the csam scanning is an attempt to establish a bit of preemptive control

Mike: but

Chris: what we got insights into since the i don't know the last episode or so it really shows us that every like worst case scenario we've we've spun in our heads about this is absolutely true and vindicated because if if this epic lawsuit does nothing else we are getting some great emails that are truly revealing the actual nature of these companies yeah i just want to take a lap

Mike: here and say so when i was talking about this crap in like 2013 2014 maybe even all the way back to 2012 people said i was crazy or whatever i was right the whole damn time thank you

Chris: it might have even been worse than you were saying you know it's actually worse

Mike: than i thought it was yeah yeah yeah

Chris: yeah that's the thing um like so the emails that are coming out and there are some great summaries we'll have linked in the show notes uh basically show that google had a secret initiative originally called project hug that offered certain very special high-profile software developers very special treatment including maybe not having to even kick back the 30 percent fee google's larry page told steve jobs in 2010 that quote there will always be places we compete and there will always be places we cooperate after another meeting between apple and google and senior executives they had notes that showed that the executives all agreed quote our vision is that we work as if we are one company that's some duopoly right there mike and that's just the tip of the iceberg like there's there was just emails that said we're not going to feature our competitors in the app store uh there is all kinds of nastiness in there a lot of them are buried in pdfs too

Mike: my impression of this and the argument i even made back in like 2014 was that this is not necessarily evil or it wouldn't be evil if apple didn't go so far and i'm gonna pick on apple a little bit to say that they didn't do this right because if you if you look at all this stuff it's actually businesses operating as businesses do and saying okay there's this potential you know developer partner or whatever that has a big enough audience that we need them and therefore we can't extract as much revenue out of them so we have to make a deal that's just like how normal business goes right remember software etc which might not some people might not be old enough listening but like they used to make these kind of deals all the time right they'd get a better cross promotion with a publisher of a game or with a software suite you know retail stores do this kind of crap all the time too the difference is apple for whatever reason wanted to have this i don't know to like wrap themselves in the in the flag of small business and like equality for the little guy well it's obviously the best

Chris: possible message they can drive out of that app store right yeah but not if

Mike: it's found to be just like completely false and it these emails reveal not only is it false but it was never in fact true right it's always been the app

Chris: store of big business right it's always

Mike: been well if you're netflix or microsoft you know but even okay like i have a surprising well i think you'll find it surprising but why don't we cherry-pick some cases here and like just for folks who maybe didn't stay up all night and read all the emails

Chris: no are you thinking like the big fish circumstance like which one are you

Mike: thinking of i was i was thinking yeah the uh

Chris: the i think i titled it sucks to be big

Mike: fish games yeah someone from app review is writing to their their upper management being like i don't have a reason to reject them are we just going to tell them nothing it's amazing it ties into the other thing there's an email and i think we link in the show notes where the app review executive is writing to the up higher-ups and saying if we're going to have an unwritten rule that you cannot compete with us to be in the app store then we should just say that right because i ultimately a lot of big fish games a lot of these come down to you can't be competitive with apple yeah

Chris: or they're gonna take there's gonna be some sort of there's gonna be some sort of down low quiet punitive action that they won't maybe even tell you about and then you as a developer just keep slamming your head against the wall well

Mike: they have multiple ways of doing this right um in the case of big fish games they just didn't like two years of trying to get on the app so they didn't let them on the app store

Chris: and i think a more interesting case is

Mike: actually like spotify or uh

Chris: before they bought shazam when shazam launched a new app to replace and the quote replace ios music player pissed them off pissed apple off eddie q writes in he forwards the their announcement uh and this and he says then matt fisher writes the team is asking whether we should promote this today the ifund guys have also called us asking if we're going to promote it today and of course then eddie replies to that email and says no promotion we are not going to promote something that puts its goal as replacing our music player unless it's significantly better than our player and this is not now there is a unless it's better than our music player which makes it sound like if it was they would consider it but something tells me something tells me especially now that music is part of their services revenue they would never consider doing that

Mike: right there's also a tidbit in there that's a little inside baseball but back when i was doing a lot of startup app development for folks let me tell you something being in the ifund versus not being in the i fund was a huge deal forget about the money itself about access and cooperation from apple i think we talked about this a couple years ago where i would have one customer who was like in the iphone they were startup they could literally call up a rep and get help with whatever was going on versus like a regular business would just be like it's a black box you submit it but you get whatever back

Chris: i'm reading through the uh the netflix stuff too where they where they're negotiating lowering netflix's in-app purchase price and talk about what netflix think would happen regarding you know subscription cancellation cancelizations and how they could possibly leverage that but they say they feel like you know for netflix they're gonna have to give them a deal here because you know they're a special customer they're it's um it's it's really something to see these emails this one's from peter stem to phil schiller and they got it right here epic got it i got him it's good to be apple right like they got they became a platform king just at the right moment right they really nailed that app store thing they almost blew it and then they really turned that around and got themselves to essentially be the gatekeeper for so so so many users i

Mike: have some bacon oh really tell me a

Chris: little bit about this bacon is it uh smoked

Mike: yes this is smoked um this is gamified bacon i think there's one thing that i i don't know that's gotten enough attention uh during this epic hearing the disclosure is basically well not basically the outright proved that games on ios are subsidizing the rest of the app store yeah if we look at these kind of like bigger rejections like the one where microsoft tried to like get on their heels and fight apple a little bit it's all about cloud gaming and i i have been dunking on cloud gaming for several years on this show but i think we're getting to it now and that is a big big market and a lot of that sweet sweet revenue that monthly recurring revenue and that's what it's all about

Chris: and how how apple handles that will be interesting and what are they going to do when all these guys switch to a web app will also be interesting what are your thoughts on on google i mean how about this name how about this [ __ ] name project hug how about that it's almost like the poor

Mike: imitation of apple's tactics in a lot of ways it's just even more ham-fisted it's like

Chris: they they they came up with a cute sick twisted name for their cutting deals with special interest to like you know pay their 30 commission or something like they got a name for it project hug it's gross like that's just it's not apple's not the only bad actor here no not at

Mike: all i mean there's stuff from from this hearing uh from this case that we've seen about microsoft and sony too they're all doing this like if you are a big enough name that they need they'll just make different deals with you i i sort of find the apple case more egregious because they're the ones who got on the soapbox you know they always have those crappy like human interest stories at the beginning of wwdc right and no offense to the individual developers or the idea that you know we should promote less certain developers i would love to be promoted too but it's that's just not the reality of the situation yeah having said that i think this is all good news in what way because once you get caught you can't get away with it so like i'm thinking back to some of the hate mail that i in particular used to get from the show when people would say oh it's just whatever you know you know second place weeps i don't know kind of kind of that apple wouldn't do that it's a fair market blah blah blah they denied it they said they didn't well now they just got caught and i think the most common

Chris: thing we got was like apple's doing this for safety of the consumer apple has to do this for the device security right

Mike: and it's [ __ ] if you read the emails it's revenue right it's it's always has

Chris: been from like the very early days from the day one

Mike: right so the the thing is once you get caught especially in i mean i don't think congress is actually going to be effective because yeah but i think apple's going to think twice and all these vendors are going to have to think twice about how how just how hard they laid the boot down on the neck of the indie developer

Chris: i wonder you know we've covered over the years these developers who have had a big enough clout to actually generate some sort of press coverage about their mistreatment and there's been dozens of them now and you wonder if they hadn't just screwed those up if they had if they had just done right and just changed a few things like as time went on they dropped the commission down and you know if they cut a special deal with somebody like netflix maybe they they institutionalize it and they make it a system that anybody can apply for and they had just done things like that over time

Mike: none of this

Chris: none of this would be happening

Mike: they can't right they're publicly traded companies the chart has to go up into the right you can't just make less money than you could make that's a good point

Chris: and especially when you have a lot of pressure to always be making more money it really does get you to the point where eventually you have to start making some kind of compromise and if you can't sell you know if you can only you can only sell so many devices there's only so many people in the market and then you got to start you got to start milking that existing market for revenue growth it's the only choice they have

Mike: let's not forget right in their little services pie which is not that little very big the largest part of it is in fact app store commissions yeah yeah that's true i i i wonder how stuck they are though

Chris: because some of these documents seem to suggest that apple and i would assume google struck deals with some of the carriers that if the app is purchased while on lte or you know on their on their on the carriers network that they're getting like 20 of that 30 percent like the carriers are getting a ginormous portion of that transaction and i don't know if that's still the case and i don't know which carriers well

Mike: grand ad monopoly has got to get in there too right but that makes me wonder

Chris: then well how much room does apple have to move you know if if they're if they're paying out 15 to 20 percent of their 30 cut to verizon or whatever they may not have a lot of room to move down they may be kind of stuck by their agreements that they stuck with these carriers to get them to allow people to download 200 megabyte apps on the carriers networks you know because when the iphone first came out remember they really limited that that was a big deal and then and the network struggled they struggled to keep up with the data use of iphone users

Mike: i still believe that the bigger problem is actually the control and the review of the applications and how restrictive they are when you are not them right again going back to spotify for a long time and i think still presently there are just things that apple music which i am an apple music subscriber i use it every day it's fine but you know if it didn't come with like everything else i have spotify is definitely the superior music service right podcast player but good music service but just that you know i can talk to my car easily and have it play apple music is i don't know like the home pods that somebody convinced me to get that i love now apple music's easier right i don't know i feel like we were at one point on this way to a glorious future of the open internet and like web applications and it was the 90s it was going to be great and then somehow facebook and the app store became the internet and now we're all just a little sad i say it's the

Chris: consumer's fault because the consumer couldn't figure out how to find without these curators you know discovery has been a problem for all types of media and and apps and these app stores and these platforms like facebook came along and they offered news feeds and they offered curated apps and they said go here you no longer have to think about it and the consumer said oh yes please yes thank you thank you and now we're in this position yeah but

Mike: i feel like you know great granddaddy was making the same argument about those newfangled magazines when they came out

Chris: probably true uh mr dominic is there anywhere you want to send people this week before we get out of here uh yeah follow me at dumaku

Mike: on twitter and i am still hiring for a qa person and a pythonista so and what is

Chris: the best way for them to find you because we had because i think we had mentioned they could like they could go find the job posting online but then somebody emailed is like i don't know how to email mike i don't know how to get yeah it's a zip recruiter slash the

Mike: mad botter inc and actually if you just google the madbotter jobs it comes right

Chris: up there you go and we'd love to have you join us live we do this show monday at noon pacific 3 p.m eastern over jblive.tv links to everything we talked about today at coder.show 428 including those incriminating pdfs you'll find our contact form there we'd love to get your feedback it was a huge part of this week's episode and we'd like to make it a big part of next week's episode too go subscribe maybe share the show with a friend who might enjoy it word of mouth is the best way to spread these here podcasts but that does do it for this week's episode of the coda radio program see you right back here next week

None: [Music] you

02_AssemblyAI.md

Chris: This is Coder radio Episode 428 for August 23 2021. Hello, friend, and welcome in to Jupiter Broadcasting's weekly talk show, taking a Pragmatic look at the art and the business and the whole world of technology around software development. This episode, it's brought to you by Cloud Guru. You know, cloud guru has the cloud playground. That means Azure AWS. And Google Cloud sandboxes on ACG's credit card, not yours. So go get certified. Get hiregetlearning at a cloudguru.com. My name is Chris, and joining me, like the podcast pro that he is, it's our host, Mr. Dominic. Hello, Mike.

Mike: Hello, Mr. Fisher.

Chris: Podcast time. Yes, here we are. I am excited to be in Colorado Springs. We win eight days off grid, no hookups, running off solar. And so now it feels luxurious to be in a campground parked. Parked and plugged into power and, like, have air conditioning and stuff. It's like, I'm fancy.

Mike: All of a sudden.

Chris: However, it's like a downtown city here. This campground is so freaking crazy busy. So apologies if I've got weird sounds in the background because there's a whole party going on around here. Are you fired up about the podcast? Are you ready to go?

Mike: I've been looking forward to this. I've been waiting for this one a long time.

Chris: Yeah, well, I mean, I kind of been the opposite. I've kind of been dreading it because the inbox this week the inbox Mike. It was so bad. We got so many emails, but good. It was a good problem to have. And so we thought we should definitely try to get to some of these. And listener Chris, not me, but listener Chris was fired up about crystal. He says, hey, guys, I'm a little bit behind, but I just listened to the Ruby in the rough episode, and I have to plug crystal at Crystal lang.org. They released version one not that long ago. It's written like Ruby, but compiled down and fast, like C. I've started to rewrite some of my Ruby in crystal, and it's been a huge success. Thanks for what you do. And sorry I haven't chilled crystal before. I'm just now catching up. Okay. So I really didn't know what else to say other than I guess, thanks, listen to Chris for firing an email about crystal. And it was totally off topic, but felt like we should include it at least.

Mike: Well, it's actually not that off topic. If you remember the dark days of the coding challenges, it was one of the languages that Wes and I used to torture each other.

Chris: Yes. Right.

Mike: If I had done my homework, I would have been able to find the episode. But unfortunately, fortunately, there's just a lot of quota radio out there in universe, and I could not find it. But it's there somewhere.

Chris: We need some sort of system that goes through and indexes of the speech, you know, like does a transcription and then generates keywords.

Mike: Why is there not like a Google search of podcast content.

Chris: They were playing around with it for a while, and then they just sort of dropped it.

Mike: Really?

Chris: Yeah.

Mike: Because I wanted to reference something for today's show from ATP. St. John Sergey said, and after an hour and a half of trying to figure out where he said it and which episode, and thank God they used Chapter markers, I still could not find it because it turns out I was a year off in my looking, and I was like, I tell it this, I'm not doing this.

Chris: Yeah, there probably are a few other services out there. There are companies out there that are doing it to monitor when brands are mentioned because it's a smaller problem scope. Right. They can just monitor for Netflix.

Mike: Sure, makes sense.

Chris: So there are businesses around that, so I think it'll expand from there. All right, well, Jim wants to know more about C Plus Plus deeds from you. He says last week, Mike tossed out one of his classic throwaway, saying that he's running C-L-L on his back end, and well, are we just supposed to ignore this? Please share as much as you're able. What motive running what libraries? What inner motivation led you to make this choice? Are you running? C and CGI and WebSockets dark matter. C plus plus. Developers are slowly steering their organizations away from cute native apps, and I want to know what you're doing.

Mike: Yeah, so definitely not cute, right? This is the server side stuff. I mean, very simple, very kind of pure anti C Plus Plus here, just C. Make the ID. Some of us are using Client. I'm using Vs code because I've just drank that Koolaid like it was going out of style. The way we're using it is maybe less web servery than either I made it sound or that you got the impression of it's more. We have Python, so the stack is like, okay, we have an application. Most of it the interaction, the Webby stuff isn't flask for the most part, and anything performance is in C Plus Plus. It was going to be Rust, interestingly enough, but finding external help for Rust is way too hard and way too expensive right now. So we kind of fell back to C. Plus plus. I have to say we have taken, I think, a little bit of a weird approach. So, one, because we control everything. This is our own thing. It's the newest standard. C plus 20. So we're using all the fun little new bits, right? The smart pointers, all that good stuff. We're also not using Boost, which I think is important to mention. We're kind of doing things just as they are. Standard library. Everything else is kind of written ourselves. I know that smacks have not invented here, but there are reasons that I think probably over the next six months we'll talk about why for something like this, where we want to control it full stack and it really is full stack, meaning from server to hardware.

Chris: To.

Mike: Not have that level of dependency. So that kind of segues into your other point, dependency management. Other than the standard stuff you would have in C make. We really don't have many dependencies that may not be true forever and.

Chris: I.

Mike: Know there are options, but this is kind of where we need performance and don't get me wrong, where we can we're doing everything in Python, but sometimes that's not applicable. I know there's ways to compile Python down. I look at them, I get it. It's just a little too voodoo for me. Easier to write a Linux Cinema and Hazard call into it, right?

Chris: Sure. Well, thank you for sharing that. Our next email is asking for us to consider intel for a moment.

Mike: I didn't bring my tissues.

Chris: I know. Yeah and your beer. To pour one out I need a 40. Jacob Wright. Can you guys talk about intel for a few minutes? I'm with Chris. That Canoma is deploying solutions that widen the chasm with their audience. But if it was a contest, I would think intel would be able to take the cake. I've never seen a company deploy more promises than actual products. While they will likely always stay relevant for a large audience, how do you see their future unfolding? Do you think Apple's upcoming M One X or M Two processors will encourage a double digit shift from the X 86 architecture? Our thoughts on intel mr. Dominic, what do you think about Intel's future and their products? Like I was thinking about our friends over at System 76. They tell me they're working on a laptop that they want to have out in a couple of years, right? And I thought to myself, god, it could be really awesome. Metal chassis, maybe it's milled from a block of aluminum, runs Linux, has replaceable parts and then I thought, oh, but it's probably going to be an intel chip or at least an X 86 chip. And that may be a deal breaker for me because it kind of feels like buying a gas gozler in an era of electric cars.

Mike: What's the time frame here? Is it like now or is it five to ten years?

Chris: I think starting now. But the way Jacob define it. It sounds like with the M one X or the M Two. Assuming there's some good performance gains there.

Mike: And I know we get a lot of heat for this. But there's no way to I think objectively look at the situation and not say that the M One class of processors from Apple isn't a leap forward. Which necessarily means that intel needs to catch up. My broader thoughts or intel made a bet like 510 years ago that didn't pan out right and they got kind of hosed. They're not going to go away. They're a big company, they'll be fine. Can they catch up?

Chris: Sure, right.

Mike: They could have. The next good idea for processor technology, is it going to happen within the next, let's say two, three years? I don't think so. But keep in mind, Apple is not going to license their processors out. Now, you might be thinking, okay, a risk architecture, like those type of processors and maybe IoT, that's true. That's a market that I don't think intel competes in very well. Right. With your IoT devices, you're kind of automation factory kind of stuff. You're probably running Arm. And I know so far, Scifi, the company Chris Latiner went to from Tesla, chris Latinner, for those who don't know, is the guy who made Swift at Apple. That's what they do, right? They do like beef spoke risk five processors for IoT devices. That's also a competitor on the desktop. I sort of feel like that's the market all nerds care about quite a bit, but kind of doesn't matter. I don't know. Also, those rise and chips are looking pretty good.

Chris: That's the wild card here, right? Maybe AMD pulls it out, maybe intel with their Arc GPU and just really gets off their ass and actually manages to produce something. But I suspect with intel it's a corporate structural issue. I think the rod is from within and they just got complacent again and their management structure changed and now they cannot fix their problems. At least that's what I worry about. And I have to say I'm not sure yet. Jacob, really, I think I'm on board with what Mike says is I need to see where Rising plays out. I need to see what these AMD laptops with AMD GPUs in them, what they look like, because we've seen a few on the market, but we haven't seen it at the scale that we've seen the intel stuff yet. I did note when I was talking to the system 76 crew, my thought was, but it's still going to be intel, right? So it's almost like doesn't matter. And I hadn't conclusively reached that, but it's starting to feel that way and I'd be curious to know if the audience out there feels similarly. It's one thing I think, too, when you're thinking about it from a theoretical standpoint, like, oh yeah, if I'm going to go blow two grand on a laptop, you can just sit there and you can throw around what you may or may not do all the time. It's easy. But when it actually comes down to I can only buy a laptop like every four or five plus years, it's not something I do all the time. This is my one chance to get it right. I want something with a good screen. I want something with a good build quality. I want something with a good keyboard. I want something with good battery life. And I want something that has high performance and it's my actual money now that I'm spending. I find it getting harder and harder to buy an X 86 based system. Thankfully, I'm just in a wait and see mode right now. C Sharpait writes in now, he says. Anyways, I can't help but take those stack overflow surveys that you guys cover as just a chance to rag on C Sharp. Even though it's not my favorite language objective, C is a bit overheated in my opinion. C Sharp, on the other hand, is an evil virus from hell forced upon the human species. I've never encountered a language that was far more removed from what it claims to be. In the case of C Sharp, it's, quote, easy to read and easy to use, but that phrase couldn't be a better description for everything. C Sharp is not. The nastiest and most unpredictably dysfunctional code I've ever seen in my life has always been written in C Sharp. It's ugly, it's easily breakable, and you can never rely on most people to write anything maintainable. The best C Sharp can do is work. In a clunky way, I was shocked when I found out that C Sharp was 60% loved, but the more I thought about it, I thought about it's probably a blessing, really, that many programmers seem to either not remember how terrible C Sharp is or because they've long abandoned it for the browser. I never forget the nightmare that is C Sharp, especially back before stuff like Vs code existed and I had to use the old Visual Studio institutionally forced upon me. There may be worse languages objectively, but C Sharp will always have a special place in my heart. Like, as much as I hate JavaScript for spawning electron, at least electrons cross platform and is actually easy to use. In fact, many of its problems stem from being so easy to use that a lot of people can build apps that just work well enough and never had to learn good programming habits. C sharp is the worst. I don't know about that. I think there are some people out there that would strongly disagree. I know some people who desperately are in love with C Sharp and trying to figure out how to balance their real relationship with their relationship with C Sharp.

Mike: Yeah, I feel for the writer here, right? Because he's got my level of love for Objective Studio, a language that I would say the vast majority of people hate. He hates the language the vast majority of people love. But the funny thing that he mentions there that he was forced to use Visual Studio. I cared a wager. One of the main reasons lots of people like C Sharp is because official Studio I don't know, man. It's not really my bag, right? I tend to run macro pop, but if you want an IDE that does lots of stuff for you and helps you out granted, I know it adds complexity with all this weird config files, but I find any argument that is a visual Studio does not succeed in its primary mission of being a great IDE for net development. A little nonsense, and I know that's not the army he's necessarily making, but see, a lot of this stuff is about tastes and preferences, right. Why do I like Objective C? Because I'll turn this into a positive argument not to vash with C Sharp because I actually kind of like C Sharp objective C until at the very end and I know and is like in quotes because it's not dead yet. I will assert that forever. Just didn't have that many features. I'm doing a bunch of Python now and there's like a million magical ways to do things in Python, right? Because of like, different libraries. Whatever the objective is, is you can send a message to any object. We have built in these very few features. Then you can import cocoa, fine, you get some stuff there, but it's a simple language with simple concepts until you crack open the C part of it and then it's still simple, but you can go cry because you have to do a lot of memory stuff. I have a feeling that some people like C Sharp because it's a big language, lots of features, and just a ridiculous amount of abstraction, right. Where Objective C is like, hey, man, you want to do that? No problem. You can cobble together traditional C blocks and, like whatever weird NS auto release pools and object sending things we've got going on here and do some dynamic stuff in the runtime. C Sharp is like, oh, dude, we have a class for that. Yeah, just throwing away just throw an Async away and we'll figure it out. Don't worry about it. I don't know. It's kind of like easy mode, right?

Chris: Yeah.

Mike: And that's not a bad thing. It should be easy mode.

Chris: Okay.

Mike: Can I bitch a little bit?

Chris: Do it.

Mike: Doing tons of Python. Loving it. Except for both PyCharm Pro and Vs code for whatever reason, cannot do IntelliSense on the SQL Alchemy like classes. So if you have a model that inherits from things like DB dot model and I think we should just take a step back. SQL Alchemy is effectively active record or entity framework from as a guy who has a serious type of problem, as Chris can tell you, that has been killing me because just not being able to hit like dot escape or I think I have a hot dot tab and seeing the options of methods and properties. I spent like 3 hours this weekend trying to figure out why it's not working and there's just a bunch of annoyed people on stack of floor. I don't understand why this doesn't work. Right? That's the kind of quality of life thing that's annoying. And if you are a Pythonista of more of a beard than me and you can explain to me why Vs code cannot figure when I do like my model query, whatever, why the dot doesn't bring anything up, I will be very happy.

Chris: I think they prefer to be called python hounds.

Mike: I thought they were like snake charmers.

Chris: Yeah, that's right. Another topic that we covered on the show that got some feedback and we're still the door still open on this one is transitioning into a manager from a developer to a developer manager. Jacob wrote in and said, there's one thing that I really recommend is sit down with each person and do a one on one session and ask them this question. Tell me about a mentor that impacted your life. He said, you'd be surprised at the kind of answers you get and insights you get into what motivates people. And then Peter wrote in with two book recommendations the Unicorn Project and The Effective Manager. We'll have those linked in the show notes. He says they give you actual advice and not just theory, and that The Unicorn Project gives you a good idea of like, how to run a DevOps shop, but in story form. So that's the Unicorn Project and the effective manager. We'll have those linked in the show notes. If you got some of your own snakes that you have to charm, those are very useful. We have more feedback to get to, but if anything we've talked about in the show strikes an interest of yours and you want us to talk about it more or expand on something, or you have thoughts or insights, go to Coder showcontact and share those with us. I'm in Colorado Springs. As I mentioned, we just did our Denver meet up a couple of days ago and there was an incident that I got to talk to you about because there's been some demands that have been made that I need to relay to you and we got to come to some sort of decision. So there was a bit of a coder upset at the Denver meet up.

Mike: I swear to God, if you say closure no.

Chris: A coder quorum was assembled. They came up with a list of demands and then they claimed one of the bars as their territory, so that got a little weird. And then they wouldn't let any of us Linux nerds enter the bar land until they had your blessing and they refused to listen. And I told them you weren't coming because we assumed you were going to be moving and they just weren't having it. So now they're holding several MacBooks from the participants hostage, threatening to spill water on them unless you show up at a future meet up. I don't know how long this is going to go on. I think they're even threatening a hunger strike. It's pretty bad.

Mike: The one in Denver is the one I should have gone to. And by the way, just for Carl, I've gained £400, so don't mind the really big jacket coming into the system. 76 factory. It's cool I've been eating Twinkies.

Chris: Oh, man. Can I just say that the hardest thing about this meet up is the fact that I gained a bunch of weight during the lockdown stuff, and so now I have to go around like, I don't fit into my clothes. It's just the whole thing is embarrassing.

Mike: I have a bunch of JB shirts that I no longer fit into, like the old coder ones.

Chris: I'm like, those are my Aspirational shirts now.

Mike: Yes, the coder like, 100 shirt is lovely. I look at it sadly all the time.

Chris: Oh, man. I know what honestly, where it really kicked off for me was a year ago we were just going independent, like this week because of the week we went independent. Like, tomorrow was the day we announced it.

Mike: Happy anniversary.

Chris: Thank you. And the many months that led up to that, the negotiation process was extremely stressful. And it was right at the beginning of the lockdown and all of that, and I just put on a ton of weight and I realized, oh, stress really does make a difference on that kind of thing. I'm going to have to just take steps in the future when I'm under a lot of stress to kind of like preventive care kind of thing. But man, oh man, was that a process. And now here we are. Here we are doing the meet up and all that kind of stuff. It's been great. So almost one year to the day, as we record, by the time this episode comes out, it will have been the year anniversary. It's pretty great.

Mike: So for folks who might want to plan to go to a future meet up, is there somewhere that these are posted?

Chris: I think we're going to come up with a meet up.com replacement right now. We're still using meet up.com. Okay, but I feel like it's time to find something else. Meetup is designed around the idea that you're holding meet ups in your area. It's not designed around the idea that you're holding Meetups somewhere far away from you. Oh, by the way, Optimist Gray points out in the chat room that you can go to Notes Jupiterbroadcasting.com codartradio or just go to Notes Jupiterbroadcasting.com and you could search there for our coverage of Crystal. There's a really nice index of all of our previous shows with a special search front end. In fact, the coder audience might find actually pretty clever. So that's notes Jupiter broadcasting, then. It's quite the fancy search. Lino comcoder go there to get $100 on a new account and you go there to support this show. The year of independent code of radios has been made possible by Leonard, and we enthusiastically endorse them. They started in 2003 as one of the very first companies in cloud computing. 18 years later, they've just that entire time focused on that mission. They haven't gone crazy. They didn't get like, some weird OVC funder in there that wanted them to get into like some crazy aspect of technology. They just hunkered down and built something incredible. And as the different things in the market evolved, lenode has consistently responded and over delivered. And that's the part that I have been fortunate to witness over the last couple of years. I'll tell you something, they've added kind of more recently that when I first started using the Note. I never thought they'd have and it's bare metal boxes. I just never thought that was going to be their game. But they looked at it. They still something customers want and it's something we can do right on top of that though, they just have an easy to use powerful cloud dashboard. They have eleven data centers around the world you can deploy to. So if you need your application to be performing, if you need it accessible, they've got you covered there. They've got DDoS protection, they've got this beautifully fast network because they are their own ISP. And the reason why I want to mention this is I have set up applications that I have deployed in different data centers that come back to their New Jersey data center for Object Storage or something like that. And I was worried that maybe that wouldn't be fast enough, but it has been exceptionally well performed the entire time we were on this trip. As the crew is recording clips, both video and audio, which man, where people just like Brent was working his tail off. We would get together in the evening at the Airbnb and we would upload all of our various footage to Lenov's Object Storage. So we have one centralized place we don't need to bother with any crazy front end software. You just upload it to their object storage. Anything that can talk s three can talk to Lino's Object Storage. They have also entire one click application deployments of the entire stack, if that's what you like, or you can do it yourself. And their pricing is 30% to 50% cheaper than other major cloud providers. But with that $100 credit you're going to get when you go to leno. Comcoder, you can really kick the tires. You could try out one of their high end AMD Epic dedicated CPU systems that just kicks the butt of everybody out there. Or you could go for their $5 a month box and see how far you can push it. You'd be really surprised by that. And Lenode made our Jupiter Colony reunion road trip possible. They were there at the Denver meetup giving away prizes and some pretty special events and memories were created because of their support. And now you can get $100 to go try it out for yourself. Just go to lenode. Comcoder, get $100 on your account and support the show. Lenode. Comcoder. I was a little fired up last week and that's what got the most feedback into the show. I was having a real struggle with Nsurl Session D on the Mac. So quick recap because I'm a glutton for pain. I thought to myself, I don't really have a great computer for this road trip. I kind of went all in on my X One Carbon at the beginning of the year and frankly, I'm just not very happy with the overall performance and battery life. Other aspects of the machine I am quite happy with. I love the build quality. I think the keyboard is great. It runs Linux because I got the Fedora edition like a freaking champ. Like if you needed a mobile workstation to do some web development, to get terminal access, to check your email, to browse the web, this thing is checking the boxes all day long. However. If you are trying to do a crazy remote live stream broadcast with a dozen different applications. Including a full digital audio workstation. Including OBS for broadcasting. Including soundboard software. Edge browser. Chrome browser. Firefox browser. All running simultaneously. Jack and pipe wire. Running the background routing options. Doing real time effects processing. And then of course. Recording each individual device and track to a FLAC file in real time to the hard drive. That's the baseline before I'm even like, browsing the web, reading the news stories, responding to the chat room or launching Slack. The baseline is a lot of work for a machine. And I just said I got to bring the MacBook Pro because that's the fastest machine I've got.

Mike: Okay, so far. I'm with you.

Chris: Okay. So I tried it and I thought, how bad could it be? And I really got nailed by something called Nsurl Session Damon last time. And at the time, because I didn't really know what it was, I was in problem solving mode. It was consuming all of my available bandwidth. It was just really going to town. First it consumed all my downstream and then it consumed all my upstream and it happened for a period of three days. It was really getting to like by the time I've done coder, I was three days into this being a problem. What I learned after the show is that Nsurl Session Damon is actually kind of a good idea in principle. I think it's actually the implementation that is the problem. It's a macOS system process that is used for downloading and uploading data for any application. Initially I thought it was like just an Icloud thing or something just for macOS, but now I understand it's actually a class that is made available for any application and any application can delegate transfers to that daemon. And the nice thing is that it will continue even after you quit the application. So you can be like telegram and say, hey, I need to download this thing and it's URL session will do it for you. The main telegram process is left free, you can quit it. The download continues. And through my investigation, basically by throwing lsof and looking at everything it's doing, Because Activity Monitor gave me information. But by using traditional Unix tools, I was able to determine, like, Edge uses this, telegram uses this. Of course, all the Icloud stuff uses this. Like, tons of applications are using this damn NS URL session in the background on my Mac. And that is why it was just constantly doing shit, because I had limited connectivity. And so things just sort of queued up and it was just dutifully going through the list and executing its list of things to download, consuming all of my bandwidth. And we got a lot of emails and that people said, yeah, this drives me crazy. Hans wrote and said he switched to Plasma, he switched to Katie Neon because he was just sick and tired of all the stuff that the Mac does in the background. I'm just curious to see if you agree. I don't think you've really suffered from this because you probably are somewhere either with high bandwidth or something. Like, you've never been plagued by this problem, I don't think. I've never had. This problem could happen to anyone. It could happen to any Mac user. You just kind of need to be like in a low bandwidth situation for an extended period of time. And there are a lot of options that came in. Nobody really wrote in with a way to use Launch Control to just disable this thing, which is what I actually expected because that was the route west and I took, was like, well, let's use the system application Launcher to just shut this thing down. But no, no, what people suggested I do is use a hammer. And I ended up doing just that. So I checked out two applications that were written into the show. One is called Lulu by a company called Objective CSE. Yes, I know you like that. And this will have a link in the show. Notes is a really neat open source firewall. It watches for outbound connections and then prompts the user if they want to allow or block that connection. And you can say temporarily block it or block it forever. And it runs at the network level using the newer APIs in Mac OS to make sure that only the software you allow goes out. And the UI is fine. It's not phenomenal, but it's fine. And it absolutely gets the job done. It's at objectiveseecom and it's Lulu. The only problem I have with Lulu is a couple of applications didn't work, like Transmit and by Word and a few other applications just weren't working. And I knew about little Snitch, which is like the OG firewall. I legitimately may have used Little Snitch back in the Mac OS nine day or eight day, a really long time ago. I'd use Little Snitch, so I thought, let's go see how they've done. Now that company's name is Objective Development. So you've got objective CSE and objective development.

Mike: Excellent company names.

Chris: Both I know and Little snitch is some big, but like the good kind. It's phenomenal. I mean, just as far as like an application design, it gives you like a real time NORAD missile map of where all your connections are going, a beautiful way to see which ones you've allowed or disapproved. It gives you prompts that it's aware of certain third party applications. If you are another independent software developer and you deny that application's connection to the Internet for some of them, like say, bartender, it'll come up with a little disclaimer, says hey bro, just so you know, these are good guys. This is just the auto update. You could just go into the preferences and turn that off and still allow this thing access to the network if you want. Like they vetted certain independent third party Mac apps and give you additional context when you go to deny them access, which is kind of nice. They've also made it really simple to understand what stuff is icloud stuff and what stuff is macOS stuff. And I just use this thing, like full stop, to just shut down as much possible on the Mac while I'm recording. And at this point in time, I've nearly blocked 60 different processes from connecting to the Internet. It is mind boggling how much crap on the Mac connects to Apple. I mean, you click the Safari URL bar, four different background processes connect to Apple. You launch Spotlight, three different processes connect to Apple constantly all the time. I mean, the list is like I said, it's 60 different processes long now that come up when I either launch an application or boot Mac OS that all phone in to Apple. And it reminds me when the Linux community cracked their pants over the Canonical dashirch and how it sent a ping to Canonical when you would search for Amazon items. Because literally every move I make on Mac OS, some process is connecting to Apple or some just automated sync process is connecting. And it is really eye opening to see how much this Mac is talking to the mothership whenever you're doing anything. And so I've just blocked all that crap while I'm recording. I've created a profile that's like a hardcore profile that tons of stuff is shut down hard and now it's working great. I don't know if it's going to cause any long term damage to my Mac, but in the meantime, I can at least use my Internet.

Mike: It's not your Internet or your computer. You need to accept that.

Chris: Apparently it feels like half ass on Apples part here because they've created this NS URL session. Damon, that kind of seems like a good idea, but wouldn't the next obvious step be to meter that in some way to not abuse a connection? It's not just like the next obvious thing.

Mike: So it's server based software and it's Apple.

Chris: True. I mean, when you look at all the stuff that just connects to the Internet, anyways, it's clear they give zero craps about the amount of network activity and network connectivity they use.

Mike: They don't give us they have no concept of like, I'm using this on a tethered connection. Granted, there is an API now to tell that you're on a tethered connection and responsible developers could act accordingly. Right, so we're able to ascertain like what exactly they're doing or it's just telemetry, right.

Chris: In most cases it looks like it's checking to see if something is on a denialist, checking to see if something is signed, or it's checking to see if something needs to sync. Or like with Spotlight, it's pulling in market information for stock searches and all this crap that I just don't need. It's preemptively caching a lot of that stuff in the background or there's a lot of assistant services and siri things that just try to connect and be preemptively ready when you click a URL bar. All that stuff though, generates metadata. Even if it's not intentional tracking, it is metadata that gets triggered. And even if it's not Apple that logs it, your ISP could be logging it. So I just turned all that crap off?

Mike: Yeah. That's not good. It's just not good.

Chris: And it's kind of a garbage experience going through it, because for like the first two days, you just live with tons of prompts about approving or denying software connectivity and you just have to wait because you can't really proactively go through and do it. You kind of just wait until something prompts and then you tell it no. And so for like two days now, I'm just sitting here saying allow or deny? And I think I've caught most of.

Mike: It at this point, but it takes.

Chris: A while to get it set right. And this whole thing could just be avoided if I could just go under the WiFi menu and say low bandwidth mode and then problem solved. Because they've already done the hard work of consolidating a lot of the background processing downloads into a session. So it's like, just give me the damn option to just throttle that thing. It could just be a little checkbox in the Wi Fi drop down that says limited connectivity or something. For goodness sake.

Mike: There will be no check boxes and drop down menus on vacuum.

Chris: No, you can't have that. Maybe at best I could get something if I held down the option key. Anyways, I really liked little snitch. I paid the $47, even though after this road trip, I don't know if I'm going to ever need it again. But I figured, you know what? Great software. Really well designed. It's been around forever and now I actually feel like this Mac is under my control and it shifts my feeling about Mac OS. I'm happier with it again. It's doing the job for me once again. And as long as I just use these tools to brute force it to behave the way I want it's. Fine. Oh, man. At least it's really cheap hardware and doesn't have a lot of ongoing costs. Datadog. Comcoderradio go there and sign up and get a free trial. And if you create one dashboard, you'll also get a free t shirt. Yes. This episode of Coder Radio is sponsored by Datadog. They are the monitoring and security platform for developers, security, and ops teams in the cloud age. You can truly unify your metrics, your traces and your logs all in one place. That means you can troubleshoot issues faster and break down silos between teams. I was just visiting with a shop here in Denver recently that uses Data Doc, and they had very positive things to say. You can create real time dashboards for over 450 different business technologies in just minutes. You can easily pivot from like a high level overview of your environment, zoom in, zoom and enhance down, get granular, and visualize the metrics of specific events. Datadog offers infrastructure monitoring, security monitoring, user monitoring, and it all brings it into one place and makes it easy for you to communicate, display, and just review. So go get started with a free trial today at datadog. Comcodradio. Go see why thousands of companies trust Data Dog as their monitoring solution. Go see why so many Coder Radio listeners are using datadog. It's datadog. Comcoderadio. If you start the trial and create one dashboard, datadog is going to send you a free t shirt. Who doesn't love free swag? Datadog.com codaradio. So there's a meta topic you and I have been discussing for a while, and it's platform control, and we've come at this in several different ways. We've talked about it in terms of app stores before. I think you could even say the CSAM scanning is an attempt to establish a bit of preemptive control. But what we got insights into since, I don't know, the last episode or so, it really shows us that every, like, worst case scenario we've spun in our heads about this is absolutely true and vindicated, because if this epic lawsuit does nothing else, we are getting some great emails that are truly revealing the actual nature of these companies.

Mike: Yeah, I just want to take a lap here and say so when I was talking about this crap in 20 13, 20 14, maybe all the way back to 2012, people said I was crazy or whatever. I was right the whole damn time. Thank you.

Chris: It might have even been worse than you were saying.

Mike: It's actually worse than I thought it was.

Chris: Yeah, that's the thing. The emails that are coming out and there are some great summaries we'll have linked in the show notes basically show that Google had a secret initiative originally called Project Hug that offered certain very special, high profile software developers very special treatment, including maybe not having to even kick back a 30% fee. Google Larry Page told Steve Jobs in 2010 that, quote there will always be places we compete and there will always be places we cooperate. After another meeting between Apple and Google and senior executives, they had notes that showed that the executives all agreed, quote, our vision is that we work as if we are one company. That's some duopoly right there, Mike. And that's just the tip of the iceberg. There was just emails that said we're not going to feature our competitors in the App Store. There is all kinds of nastiness in there. A lot of them are buried in PDFs too.

Mike: My impression of this and the argument I even made back in 2014 was that this is not necessarily evil or it wouldn't be evil if Apple didn't go so far. I'm going to pick on Apple a little bit to say that they didn't do this right. Because if you look at all this stuff, it's actually businesses operating as businesses do and saying, okay, there's this potential developer partner or whatever that has a big enough audience that we need them and therefore we can't extract as much revenue out of them. So we have to make a deal that's just like how normal business goes. Remember software, et cetera, which some people might not be old enough listening, but they used to make these kind of deals all the time. They'd get a better cross promotion with a publisher of a game or with a software suite. Retail stores do this kind of crap all the time too. The difference is Apple, for whatever reason, wanted to have this, I don't know, to wrap themselves in the flag of small business and like a quality for the little guy.

Chris: Well, it's obviously the best possible message they can drive out of that App Store, right?

Mike: Yeah, but not if it's found to be just like completely false. And these emails reveal not only is it false, but it was never in fact true.

Chris: Right. It's always been the App Store of big business, right?

Mike: It's always been, well, if you're Netflix or Microsoft surprising, I think you'll find it surprising. But why don't we cherry pick some cases here and just for folks who maybe didn't stay up all night and read all the emails, are you thinking.

Chris: Like the big fish circumstance? Like, which one are you thinking of?

Mike: I think I titled it Sucks to Be Big Fish Games.

Chris: Yeah.

Mike: Someone from Africa is writing to their upper management, being like, I don't have a reason to reject them. Are we just going to tell them nothing? It's amazing. It ties into the other thing. There's an email and I think we will get in the show notes where the App Review executive is writing to the higher ups and saying. If we're going to have an unwritten rule that you cannot compete with us and be in the App Store. Then we should just say that because ultimately a lot of Big Fish Games. A lot of these come down to you can't be competitive with Apple. Yes.

Chris: Or there's going to be some sort of down low, quiet, punitive action that they won't maybe even tell you about. And then you as a developer just keep slamming your head against the wall.

Mike: Well, and they have multiple ways of doing this, right? In the case of Big Fish Games, they just like two years of trying to get on the App Store and let them on the App Store. And I think a more interesting case.

Chris: Is actually like Spotify or before they bought Shazam, when Shazam launched a new app to replace, and the quote, replace iOS music player pissed them off, pissed Apple off Eddie Q rights. And he forwards their announcement. And then Matt Fisher writes, the team is asking whether we should promote this today. The iPhone guy have also called us asking if we're going to promote it today. And of course, then Eddie replies to that email and says, no promotion. We are not going to promote something that puts its goal as replacing our music player unless it's significantly better than our player. And this is not. Now there is a unless it's better than our music player, which makes it sound like if it was, they would consider it. But something tells me, especially now that music is part of their services revenue, they would never consider doing that.

Mike: There's also a tidbit in there that's a little inside baseball. But back when I was doing a lot of startup app development for folks, let me tell you something. Being in the I fund versus not being in the I fund was a huge deal. Forget about the money itself, about access and cooperation from Apple. I think we talked about this a couple of years ago where I would have one customer who was like, in the iPhone, they were start up. They could literally call up a rep and get help with whatever was going on versus like a regular business would just be like, it's a black box. You submitted it, but you get whatever back.

Chris: I'm reading through the Netflix stuff, too, where they're negotiating lowering Netflix's inapp purchase price and talk about what Netflix think would happen regarding subscription cancellations and how they could possibly leverage that. But they say they feel like for Netflix, they're going to have to give them a deal here because they're a special customer. It's really something to see. These emails. This one's from Peter Stem to Phil Schiller, and they got it right here. Epic got it. They got them. It's good to be Apple, right? They became a platform king just at the right moment, right? They really nailed that App Store thing. They almost blew it. And then they really turn that around and got themselves to essentially be the gatekeeper for so, so many users.

Mike: I have some bacon.

Chris: Oh, really? Tell me a little bit about this bacon. Is it smoked?

Mike: Yes. This is smoked. This is Gamified bacon. I think there's one thing that I don't know that's gotten enough attention during this epic hearing. The disclosure is basically not basically outright proved that games on iOS are subsidizing the rest of the App Store. Yeah, if we look at these kind of like bigger rejections, like when Microsoft tried to get on their heels and find Apple a little bit it's all about cloud gaming. And I have been dunking on cloud gaming for several years on the show. But I think we're getting to it now. And that is a big market and a lot of that sweet revenue, that monthly recurring revenue and that's what it's all about.

Chris: And how Apple handles that will be interesting. And what are they going to do when all these guys switch to web? Apple will also be interesting. What are your thoughts on Google? How about this name? How about this name? Project Hug. How about that?

Mike: It's almost like the poor imitation of Apples tactics.

Chris: In a lot of ways it's just even more ham fisted. They came up with a cute, sick, twisted name for their cutting deals with special interest to pay their 30% commission or something. Like they got a name for it project Hug. It's gross. That's just it. Apple is not the only bad actor here.

Mike: No, not at all. I mean, there's stuff from this hearing, from this case that we've seen about Microsoft and Sony too. They're all doing this. Like if you are a big enough name that they need, they'll just make different deals with you. I sort of find the appliques more egregious because they're the ones who got on the soapbox. They always have those crappy human interest stories at the beginning of WWDC. There's no offense to the individual developers or the idea that we should promote lesser developers. I would love to be promoted too, but that's just not the reality of the situation.

Chris: Yes.

Mike: Having said that, I think this is all good news.

Chris: In what way?

Mike: Because once you get caught, you can't get away with it. I'm thinking back to some of the hate mail that I particularly used to get from the show when people would say, oh, it's just whatever, second place sweeps, I don't know, kind of bull that Apple wouldn't do that, it's a fair market, blah, blah, blah. They denied it. They said they didn't. Well now they just got caught.

Chris: And I think the most common thing we got was like Apple is doing this for safety of the consumer. Apple has to do this for the device security.

Mike: And it's bull. If you read the emails, it's revenue. Right.

Chris: It always has been from the very.

Mike: Early days, from the day one. Right. So the thing is, once you get caught, especially, I don't think Congress is actually going to be effective because yeah, but I think Apple is going to think twice and all these vendors are going to have to think twice about just how hard they laid the boot down on the neck of the indie developer.

Chris: I wonder we've covered over the years these developers who have had a big enough clout to actually generate some sort of press coverage about their mistreatment. And there's been dozens of them now. And you wonder if they hadn't just screwed those up, if they had just done right and just changed a few things. Like as time went on, they dropped the commission down, and if they cut a special deal with somebody like Netflix, maybe they institutionalize it and they make it a system that anybody can apply for and they had just done things like that, over time, none of this would be happening.

Mike: They can't. Right. They are publicly traded companies. The chart has to go up into the right. You can't just make less money than you could make.

Chris: That's a good point. And especially when you have a lot of pressure to always be making more money, it really does get you to the point where eventually you have to start making some kind of compromise. And if you can't sell, you can only sell so many devices. There's only so many people in the market, and then you got to start milking that existing market for revenue growth. It's the only choice they have, let's.

Mike: Not forget right, in their little services pie, which is not that little, very big. The largest part of it is in fact app store commissions.

Chris: Yeah, that's true. I wonder how stuck they are though, because some of these documents seem to suggest that Apple and I would assume Google struck deals with some of the carriers, that if the app is purchased while on LTE or on the carriers network, that they're getting like 20% of that 30%, like the carriers are getting a ginormous portion of that transaction. And I don't know if that's still the case, and I don't know which.

Mike: Carriers well, granted, monopoly has got to get in there too, right?

Chris: But that makes me wonder then, well, how much room does Apple have to move? You know, if they're paying out 15% to 20% of their 30% cut to Verizon or whatever, they may not have a lot of room to move down. They may be kind of stuck by their agreements if they stuck with these carriers to get them to allow people to download 200 megabyte apps on the carrier networks. Because when the iPhone first came out, never, they really limited that. That was a big deal. And the network struggled. They struggled to keep up with the data use of iPhone users.

Mike: I still believe that the bigger problem is actually the control and the review of the applications and how restrictive they are when you are not them. Right. Again, going back to Spotify for a long time, and I think still presently there are just things that Apple Music, which I am an apple Music subscriber. I use it every day. It's fine. But if it didn't come with, like everything else I have. Spotify is definitely the superior music service, right? Podcast player, but good music service. It's just that I can talk to my car easily and have it play Apple Music. I don't know, the HomePods that somebody convinced me to get that I love. Now. Apple Music is easier, right? I don't know. I feel like we were at one point on this way to a glorious future of the open Internet and web applications. And it was the was going to be great. And then somehow Facebook and the App Store became the Internet, and now we're all just a little sad.

Chris: I say it's a consumer's fault because the consumer couldn't figure out how to find without these curators. Discovery has been a problem for all types of media and apps. And these app stores and these platforms like Facebook came along and they offered news feeds and they offered curated apps and they said, go here, you no longer have to think about it. And the consumer said, oh, yes, please. Yes, thank you, thank you. And now we're in this position.

Mike: Yeah, but I feel like Great Granddaddy was making the same argument about those newfangled magazines when they came out.

Chris: Probably true. Mr. Dominic, is there anywhere you want to send people this week before we get out of here?

Mike: Yeah, follow me at jumanuku on Twitter. And I am still hiring for a QA person and a Python Easter.

Chris: And what is the best way for them to find you? Because I think we had mentioned they could go find the job posting online, but then somebody emailed in like, I don't know how to email, Mike.

Mike: Yeah, it's ziprecruiterthemadbottering. Actually, if you just Google the Madbutter Jobs, it comes right up.

Chris: There you go. And we'd love to have you join us live. We do this show Monday at noon, Pacific 03:00 p.m. Eastern, over Jblive TV links to everything we talked about today at Codr show 428, including those incriminating PDFs. You'll find our contact form there. We'd love to get your feedback. It was a huge part of this week's episode and we'd like to make it a big part of next week's episode, too. Go subscribe. Maybe share the show with a friend who might enjoy it. Word of mouth is the best way to spread these your podcasts, but that does do it for this week's episode of the Coda Radio program. See right back here next week.