PIT Reflections: A Million Children's Recitals
Aug 04, 2020 21:22 PM
If you see a way to make this better, you can submit a PR on the Github Page for this Article
PIT Reflections 2020-Aug-04 I've been doing a lot in public, from talks and livestreams to appearances on other podcasts and taking interviews.
Talks: - Playing your way to Profit (The Motivational Edition) - MacStock Conference and Expo (Currently Not Available Online)
- Working with Files in PyTest - San Diego Python
What is up everybody?
Miller here back with another productivity in tech reflection.
The 1st 1 for the month of August.
Oh, boy, it has been quite a month.
Quite quite quite a month, I guess.
Technically, it hasn't been quite a month because it's only been a few days in, but July was busy, busy, busy, busy, busy, busy and we're gonna get into that.
But we're also gonna be talking about some of the other things going on around the world of pit in my world, in all of the things that I've been noticing, a little bit of news and some other stuff.
So let's let's jump right in.
Let's catch up here.
The how is July for me shall I was busy, so I was, like, stupid busy.
Ah, I did a lot of things.
In July, I gave a couple of talks, one for the Mac Stock Expo in which I talked about playing for profit, which basically was the idea of creating Elaine for yourself by having fun and then being able to financially sustain yourself from that lane that you've created.
I also gave a talk.
Ah, local meet up at San Diego Python call the Working with files and high test Um, I There's a huge bag story behind that.
But long story short, I thought that the pie test testing framework worked one way in a certain area.
But turns out it doesn't.
And I did a little deep dive into that and then basically revealed my findings in about a 12 minute talk.
But what else did we dio?
We guest on a couple of shows.
Um, neither one of them are out yet.
So your ears peeled in your eyes, appealed for you know those announcements, I will share that information on my website, as well as on Twitter and microdot blawg whenever things go live.
But, yeah, I had a blast talking with the hosts, and I don't get to be on other shows very often.
So when I am, I'm, like, super pumped on.
If I remember correctly, I don't remember the first.
But when I did the first interview, I'm pretty sure it was either in late June or early early July.
So I may have done three guest appearances this past month, which is just a lot for me.
No, something that I was wondering whether or not I should talk about publicly, but ultimately decided that I will, because I think it's important.
But I took a few interviews this month, and what does that look like?
Oh, boy, I will say, for one.
The interview process for software engineering or software development or developer advocacy.
All those things.
The process is a very, very long, then tiring, especially if you're already employed.
I am fortunate to be in a position to where I can take an interview and kind of not worry as much because I am act, you know, actively employed.
I think that more than anything, employers want to see what you have to offer, not on paper buds out loud.
And I could talk about this as a deep dive and say, like, Oh, hey, you want to be practicing out out?
I think that there's a lot of value in that.
I think there's a lot of value and letting people see the process.
In fact, that's why they do coding like coating interviews like the coding challenge stuff where they say, Oh, hey, we want you to solve this problem.
But remember, it's more about how you solve the problem.
Not necessarily that you solve the problem like they want to see the process.
They don't want to necessarily just see a solution.
So that said, I feel very fortunate because people approach to me based on the things that I was already doing.
So it became Hey, we see what you're doing.
Do you want to do that thing for our company instead of Hey, I see you're you're looking for work.
Can you explain to me why you would be a good fit?
It was like here.
We already know you would be a good fit.
Or at least we have a really good idea that you would be a good fit.
So how do we go through this song and dance of the recruiting process to one.
You know, this stuff that you're doing on a regular isn't just some flu gets.
It's something that you're able to do.
It's gonna be something that's able to be translated into the job.
But the only way I was able to have those conversations was by practicing out loud by having these conversations with virtually myself and sharing it out with the world for whoever would decide to listen.
And that's what one of the things that I talked about in my Mac stock talk was, even if you know that it isn't the best, even if you know that there aren't a ton of people watching, treat it like a Children's recital.
Allow yourself to look a little goofy, allow yourself toe, learn from those around you and take that information and publicly improve because people want to work with people who have the ability to improve.
And just because you couldn't do a thing now doesn't mean that you won't always be able to do that.
You'll be able to do that thing eventually, with enough working and maybe a little bit of advice, a little bit of documentation reading, you'll be able to figure it out.
In fact, that's one of the things about streaming at night, which is something we will talk about in a second.
That has been kind of a challenge for me because I struggle to learn new things when I'm tired, like most people.
So if I've worked, you know, full shift.
If I worked a full shift and then all of a sudden I am also now streaming and trying to perform and trying to learn all of these things at the same time.
Then I often struggle with the ability to do that when usually I wind up going to bed, not having solved the issue, waking up and then solving it within the first couple of hours in the day.
So there's something to say about.
I probably could even just flip my streams, wake up early and do like super early warning streams when I'm like first waking up, get some coffee and me and I'm ready to go.
But I don't see that working out for my sleep schedule as well as I don't see my family appreciating that very much because, you know, we have these hardwood floors.
It's an older house, they creek.
I would have basically be waking up everyone, and that's never good.
But all that to say the interview process still has its quirks.
I will say that I have been lucky in that I have had people on my team promoting the crap out of me, saying like, Hey, you know this person is an amazing developer.
This prison is an amazing person you should totally higher in like I've had.
I've had, like, a street team out there working.
And for those listening that are on that street team, thank you so much.
Even if I get hired, keep it up.
Because, I mean, I think that it's one of those things, like the more more people that air, seeing the results, the fruits of our labor, the more and more will be able to not only help me but help others that are in the same position.
I would love for people to come to me, saying, Oh, hey, by the way, are you looking for a job?
But I can say no.
But I know of some people that are looking that would make great hires based on the work that I've seen them put in.
So yeah, that's it.
Like I said, I've been going through that process.
It has been a challenging one, but a rewarding one as well.
For those that do that have been following me for a long time.
They know that Ah usually stream heart stone.
I do another podcast called Bob's Tavern Gas, where I I talk about Harston battle grounds with my friend Don.
And for the past a month, roughly close to a month now we've been streaming.
I've been streaming coating instead of harsh stone like almost every day, like we're probably 85 90% like almost every week day.
I think we've only miss like one or two days, and I was because I wasn't feeling well.
And then on the weekends, even though I normally say I'm not going Teoh, I wind up streaming anyway, so that's been fun.
If you want to catch me streaming some some python stuff for some react stuff.
Whenever I jump into that, you can do so on twitch.
I want to get well at a link to the show notes, and we'll dive into that.
But streaming is is just I don't know.
There's something about writing code and having other people be able to see your code like I don't have large audiences on Twitter, anything like that, and I'm not really trying to chase that or like gain a large audience.
But at the same time, it's great for when I do have a few people in there and I have run into like a mental block.
Especially like I said, like I'm recording this late at night.
Normally I would be streaming, but my computer decided to crash, so we decided to give it a little break.
But when I hit those walls, it's nice tohave someone in the audience to help out ask questions.
Usually, some of the audience members aren't very familiar with, like summer familiar with what I'm working on, but they're not familiar with the actual code.
So they're asking questions, and I'm explaining what I'm doing as I'm doing it, which really helps to drive in the why.
And of course, I always run into the same errors.
You know, nothing seems air is the same issues that many other developers run into in your writing code, and you're trying to have your having to explain it.
Sometimes I write code that doesn't need to be written.
I find myself doing that more than anything of like writing code that wasn't necessary, Onley to catch myself and then wind of deleting all that later and be like, well, that was dumb and moving on.
But the idea of consistent streaming is something that I've been playing with.
It's something that if I were to be a dull developer advocate, that's something that I would possibly be interested in doing.
I'm like, not on behalf of the company, but like being like Oh, hey, by the way, I'm a developer advocate.
I worked for this company, but I'm also a streamer, and we're gonna stream me learning about this thing and do that mostly to bring a little bit of attention to the coup.
Whatever company I'm working for, it would be cool to be able to build some of the things that I want to build out in public because it's gonna help me learn about whatever company I'm working with or whatever.
And then it's also going to be a way for me to connect with more and more people in the community, which is always like the job of an advocate.
See, Like I said, I've been doing podcast interviews have been streaming.
I've been, ah, interviewing all these things.
And then somewhere in there, I've been finding time to work on projects.
So the first half of the month was basically just render engine.
I was struggling to get sub collections working.
What was it was something weirdo.
I remember what it was.
I was trying to get Paige Nation working.
I had so collections working.
And then somehow when I made pagination work, I broke sub collections.
So then I had to go back and redesign that model.
And at the end of the day, it works.
I'm happy everybody's happy, and that's cool.
Then I turned my focus over to transcript er ah, which is definitely the younger of the products.
To me, it's the more you know, impactful of the products.
I have people that are actively wanting to use it.
So that's kind of what I've been shifting my focus over to it more.
But in doing that, it's been growing really quickly.
Um, we have basic functionality already done.
I'm actually working on extending that functionality to support up loading of, uh, audio files to get them transcribed, which is something that a lot of people who I demoed transcript or too like they all thought, Oh yeah, I just use your program for everything.
It's like no, actually healing used for the second half on, but I have to go through this like wild process of showing them how you do all that stuff.
But now I could just be like, Oh, yeah, No, you start.
You load this as an item, you give it a key, and then you tell it, Teoh, upload, start.
And then later on, you was pulled the data down.
So now I got to play with the idea.
And this is where we're going to get into the news because I think this is kind of important, and it's kind of relevant right now.
So for those that don't know, productivity in Tech is a podcast, a brand in a business.
The brand is always kind of loose.
It's like me helping developers create content and all that stuff.
Um, the podcast is obviously the podcast, and the brand is kind of my personal brand.
The K J.
Miller brand productivity in Tech has stronger brand recognition than K.
Miller does honestly, So I don't know why.
I just don't just own it, but I don't know.
I kind of like having my name on something, you know, it's kind of nice, but the one of the things that pit the business handles is doing transcriptions Right now.
I'm working on, um, sealing the deal on a pretty good size project that will have me working or doing a lot of transcriptions in the near future, which is actually why I started working on this as much as I did.
That project also led me to adding support for transcribing video two transcript her before it was just I worked with your audio.
I give you a transcript and we call it a deal.
Now it is.
Hey, we support sub rib, which is SRT, which is kind of like the subtitles that have been that you see on, like, DVDs and like, YouTube and all that stuff.
So now transcript or supports that.
But in working on that, the problem that I always run into is there always seems to be a hiccup when you start talking about developer focused praises as a python developer, you know, we talk about like, Jenga.
We talk about, um, trying to think of some other Gunn's sequel.
Alchemy is a good one.
Ah, post grass is another good one.
So, and working on that, I generate the transcription first using a machine transcription.
So like Amazon transcribe is what I use currently I may switch over to azure speech to text because it looks like it's like it's just better.
I think it's cheaper to, so that's kind of nice We'll have to see.
But the thing that makes transcription slow is after you get those machine generated transcriptions.
You didn't have to go through them again by hand and make any corrections, adjust lines.
All those things that make doing transcriptions more complicated.
Then one would think.
In fact, that's why transcriptions are relatively expensive, mine included.
That's also why a lot of folks don't want to transcribe their content.
Which brings me to a couple of things.
The Future of Transcript Transcript er is first and foremost a wrapper around all of these transcription tools.
It's designed so that you basically give it whatever service you want.
You tell to run.
It runs easy day.
You don't have to think about which service you want to use, which one's better or whatever, and maybe you do when I think about that.
But you can kind of switch off using the same methods and names and systems and just go for it.
But I want to follow up on that with knowledge that will allow me to analyze the work that's being done and come up with rational solutions for common errors in the system and before you message me.
Yes, I know most of these systems have dictionaries that you can input and load and all these things.
And maybe in a perfect world, we would build those dictionaries and add them to each one and solve all the problems of the world.
But here is the problem.
If I need to load Django phrases if I need to load dot net phrases dotnet terminology into my system, I have the one create that that vocabulary for each of the systems that I'm using.
Then I have toe upload those vocabularies, and every system is different.
Some of them have restrictions.
Some of them don't.
Then any time it changes.
Like those vocabulary phrases didn't have to be pulled down, updated, re uploaded, all that good stuff.
Wouldn't it be easier if I just went with what I got and use the power of quote unquote machine learning to make those changes after the fact?
And that's what I'm thinking of doing now, and that's a huge project that I would love your support on that Project one.
It's gonna be open source.
If you want to contribute code to it, that's great, too.
I'm gonna need a large data sets reaching out to conference organizers and transcribers once I have this system more robust.
My goal is to reach out to content creators who are building these transcription libraries, whether it's developer podcasters, whether it's YouTubers, whether it's course creators, whether it's, um, Conference organizer's, there will be a lot of information joining the Web.
And the gate keepers of the video industry do not seem to want to put forward tools that will help share the load of work that needs to be done.
YouTube just announced that community contributions will no longer be added to the system.
I think it's like at the end of the month I think this is a terrible decision, mostly in that many content creators do not have the knowledge to open source their transcriptions in a way that many people can make updates and access them.
But also many people don't focus on the accessibility impacts of transcriptions, and it will take someone else in the community to help with that that charge so by YouTube saying, we're not gonna let the community contribute to transcriptions or close captions.
And I know the argument is that not enough people are using the tool.
I doubted something that you would say is too expensive because you're still gonna offer the exact same.
The resource is for the content Creator.
You're not getting rid of transcriptions completely.
I don't understand why you get rid of this service.
Maybe you do.
You don't make it visible on like the front screen.
Maybe they have to go and like, go to like the more options and, you know, click it in there or something.
But I don't think it was something that needed to disappear, unless Google is like doing what they usually do, which is constantly shifting the focus of people from one area to another.
And the easiest way to do that is to remove the area of focus for them.
Currently, which leads me to my last point.
Earlier this month, there was these congressional hearings with four of the major companies in the tech space apple, Amazon, Facebook and Google.
While the conversation of antitrust in some of these conversations I thought were a pervy in Some weren't to me.
What I and more afraid of now, could be best summed up in a wonderful thought that I saw on Twitter today.
This came from Chris Hold graph on Twitter.
You should give him a follows C Hold a graph of the link in the show notes that says, Do you folks think that we're about to enter a new era of centralisation and open source?
It feels like more more tool chains, framework services and etcetera are controlled by fewer organizations.
Here's of what scares me.
Currently, I'm not trying to compete with the biggest names in transcriptions.
In fact, I'm using their open source tools and requiring that people that use transcript er provide their own credentials for them.
So if you want to use Transcript er, you have tohave.
Currently, you have to have an AWS account.
I'm All I'm doing is making making it easier to use their service.
But I already see a problem when, in terms of conferences, there are very few choices that people can go with that air affordable and know the lingo of the developer.
They're very few stenographers that could do real time, close captioning guage.
It's still amazing to me.
I wish I wish I knew how to do, um, the Stiner type like stuff that blows my mind.
I love those conversations with folks, but at the end of the day, we need more companies.
Right now, there is a huge amount of wealth in Silicon Valley that resides in the hands of less than 100 people.
That's not to say that there aren't other people out there that are really doing it.
I saw there was a wonderful block posts about Tailwind, like growing to be like a multi $1,000,000 business and stuff like that, which I think is amazing.
Um, I have my questions about why it, Selwyn, because I'm just not a fan of the u I.
But that's me that doesn't take away from the stuff that they're doing.
There have been so many conversations on this podcast on other shows about the idea of fixing the the wealth disparity in America, and I think a big part of that is reducing the amount of oneness in American technology.
Every company that is doing something great is one acquisition away from it being a feature, a feature that could be removed.
It will a feature that can be regulated and locked down and a feature that can be monetized to the extreme.
And I'm not saying that this happens with every single feature, But we've seen it time and time again, where a great idea gets swallowed by a larger company for the patents, for the licensing and for the economy.
And in doing that, we're now stifling the ability to distribute wealth amongst the many and instead that wealth remains in the hands of the few.
Now I'm not here to solve the economy problems, but what I can do is follow the mission statement That is, productivity in tech.
And that is we are a company that helps people in tech create content for people in tech and anything that would cause us to skill.
Beyond that focus is something that I'm not quite interested in, and it's something that I would question if there others out there that want to help achieve that goal.
I don't believe that it has to be everybody's mission.
I think that there's a good a good opportunity for people to help people that aren't in the tech space there some wonderful knishes that can be helped.
In fact, many of the tools that I make could be used outside of the tech space.
It just so happens that I build them with the developer first.
So a lot of the decisions that I'm making are based on that.
So I'm gonna wrap it up there with this episode.
If you want to connect with me, you can do so.
Cagey miller dot com Of course, you can follow the podcast podcast.
Stop productivity in tech dot com.
Productivity intact is all one word.
And of course, if you like render engine transcript or any of the developer works that I'm doing, you can support my work on Get Hub using to get have sponsors link in the show notes.
Now, Thank you so much for listening.
I hope that we've been productive.
I hope that we've inspired you to think a little bit about what you're doing currently, and I hope that you will challenge the thoughts of the majority of the power being in the hands of the few when it comes to the economy.
When it comes to ownership of Open source.
And ultimately, when it comes to your productivity, I would challenge you to do something that's just for you.
Do something that's gonna, I hope, solve your goal.
Don't do it for the money.
Don't do it for the accolades.
Do it because you want to learn.
Do it publicly.
Make mistakes is miss.
Result would say, Take chances, get messy.
But ultimately, that's the ticket to being more productive.
I'm Jay Millar.
This spin productivity in tech reflections Have a good day.