"It's just a button, why is it taking so long?" is a sentence capable of instantly curing low blood-pressure conditions. Only downside is the instant depression.
You frontend developers are chill. Imagine us full-stack developers, who not only have to do frontend of the button + backend api changes, but also kubernetes stuff 🤣
lol that button thing hit home. Client wants a "print" button put on a data table and added to the context menu. Ok so just print the page? No, a custom-tailored report with specific formatting and watermarks. The data they want exists across 18 tables and in 1-1, 1-many, and many-many related relationships. The data fields they want keep evolving, as well as the formatting each time they see the prototype report, as well as the report will display different data fields and be in different orders depending on certain conditions of said data. Is it a Template A? Does it contain X number of software items? Is it Tuesday!?!? 2 months later...."Why is it taking so long, IT'S JUST A PRINT BUTTON!!
Sounds like a requirements nightmare. Force them to specify the requirements, maybe they'll realize how crazy it is. At some point, you have to politely say no and make them understand why.
"You just want to add a button but Bill is on vacation" is the realest thing I've ever heard. That has been my entire experience in software, just constantly waiting for [person] to get back from PTO or vacation so that you can do the one thing that's preventing you from doing everything else you need to do.
10 years ago every engineer my work was blocked for a week because the TFS build engineer was out for a week and had exclusively locked a file. We just had to work without source control for the week I think.
@@georgehelyar During my last internship, there was a corporate merger and all the interns were completely locked out of the system for the last 10 days of the internship. This happened because they were doing a really ugly layoff technique where they deleted the log-in credentials of everyone who was getting laid off. This included the log-in credentials of every intern and they laid off the guy who was in-charge of everyone's credentials so we just had to sit on our thumbs for the last two weeks of that internship before returning the company laptops. And big tech wonders why everyone is leaving for trade jobs.
That’s the opposite of my experience in software. I do my own thing (solo dev), so if something needs to be done, I just do it. Unless I procrastinate.
I'm mainly a backend C# and SQL developer but can do frontend as well. One mid size company I worked with sales to put a "Tax Exempt" checkbox on a page. This was later in the day. Next morning she asked if the checkbox was done!!!! I said no... she said it's just a checkbox! I asked if she just wanted the checkbox there or does she want it to work also. So had to change 50+ C# files in backend code,, some reports, and some stored procedures to get it to work. Guess she thought putting the checkbox on there that the application would know what to do with it? LOL
Yeah... It's not just front end! As Full Stack, I feel like I'm getting disillusioned at ALL ends! Front AND back! Because the backend is fast catching up in being caught up in unnecessary hyper-complexity too!!! With Everything being distributed into tiny little pieces all over the cloud! With each little piece being an entire LAW unto itself!!!
The startup I worked at: "WE WANT QUALITY! MAKE QUALITY STUFF!" (They didn't tell us that he wanted a pretty rapid prototype, so I wrote software to be decently stable, and the CEO hated that). Surprisingly, the "Jira ticket for that" event almost happened when some guy in the building went to a local grocery store to buy a bunch of grapes for a snack. Middle manager was flipping the you-know-what out. "Bruh. He's gone for maybe 15-20 mins. If that's enough to derail the project, this company is screwed".
If you are a startup or a consultant, getting the thing working quickly is the only acceptable path. Write as high-quality code that you can, but don't fret about making mistakes, and if someone harasses you about it, ask them if they want something that works or is extremely high-quality and feature-rich. Pick one.
@@Joe-ku1ko True, but we are talking about a guy that can't actually say it straight out. Turns out what he wanted, with us trying to psychically read his mind, was just some UI to show off. The thing didn't even have to work! Younger me couldn't figure that out, but he could have just said it directly. Then again, if you did this, "Why isn't our stuff stable?" I actually told him that you can't expect something like that to actually work without crashing a lot, and he did not like that. You know what Prime says about 10Xers? It applied at that place. The perceived 10Xer didn't make stable stuff, but the CEO got what he wanted
@@Joe-ku1koas someone working at a consultant firm, making in towk is king. Everything else is a nice to have (except documentation, we are very careful with having everything documented)
I cried this week. Something that has been working for years was breaking in very annoying ways that could not be replicated in debug mode and the failures were the exact failures I was trying to avoid in my design. Then, I took a moment to refactor a stupidly written class (I don’t know what the was I thinking. Then again, I’m support, IT, sys engineer, project management, devops, tech lead, and lead software engineer so maybe I forgot a hat when I wrote it). The class affected my task scheduler which indirectly affected the lifetime of triggered tasks which affected whether some state was committed to disk as expected. Since I had to keep prod flowing while figuring out wth was happening, I cried every day… Software dev is truly an exercise in not crying.
4:40 as a famous developer once said: "You silly product manager! You know nothing about GALACTUS's pain, my pain. Delivering this feature goes against everything I know to be right and true, and I will sooner lay you into this barren Earth than entertain your folly for a moment longer."
That's the thing I hate about reaction videos in general. I could watch the source and save much time, provided most reactions would won't add much in general 🤷
That "net is built with old tech" part hits hard. I recently had to review a 10 years old app that somehow got a new version, but security team did not want to approve it. I checked the code: Webpack 1.0, libraries so old you can't upgrade them without upgrading / replacing dozens more, tons of libraries on top of tons of other libraries including an internal library for components that was acquired from one of the child companies and was no longer maintained. And the problem? Dep of a dep of a dep 5+ levels deep should be upgraded because string method X is bad or some minuscule shit like that. Which would require upgrading the main dep, which would require other deps changes, yadda yadda. After a day of trying to not colapse the Jenga tower I was told to write down why this can't be done, security team approved another of very many exceptions, and I went back to my oryginal project.
I felt the MTB rant. Currently am in the middle of the same thing happening in our project. It started as a simple enough web application, a .NET GUI and an Oracle DB, the DB had what passed as the application layer in PL/SQL. One webservice of some large data inputs we got once a week, that the customers then could work with on the GUI, and a DB Link for a project that ours spawned off from. Then came middle management and wanted scalability.. for an app that has maybe 50 users, country wide, max. So now we have an API between GUI and DB, and spend half a year ripping the app layer from the DB, so it can run on multiple servers, if needed (Not needed yet and probably never will be). We since added a ticket system that needs to communicate to the sister project, so got 2 APIs for that too, one for our end, and one for their end. We also needed a job service to fetch data from a third party data source, actually make that two for another one, because microservices, amiright fellas? I have 6 different .NET solutions in 3 different versions ranging from Framework 4.8 to .NET 8. Sometimes I wonder where it all went wrong.
This happens with any project. The bigger the product, the bigger the binary decision tree and the longer it takes to add even the simplest of features. Eventually, it's cheaper to make a new project than it would take to fix all the bugs in the old project.
Lol yeah but it definitely is worse in some orgs than others, I exited my startup and went to work pretty high up at an established PE firm. They had an internal platform that was ANCIENT; I was assigned to getting it up to date. The engineers there were INSISTENT that it would take months to introduce just some pretty basic ideas I had to make the UX better. They claimed it couldn’t be done. So over one weekend I rewrote the entire front end in React and ported over the backend to a Node MVC microservice. Showed it to them on Monday and said this is what they’d be working with. The team lead was batshit and started yelling at me, saying there was no way they could all learn the new stack and he wasn’t going to do it. So I fired him. He said there was no way they could get this project done without him. Surprise, surprise… two weeks later all of the features and more that I wanted were implemented and deployed to users. Not only that, but the devs were much happier. Eventually promoted one who was really eager to learn to lead, and he ended up taking one of my roles when I left. Think he’s still there.
I actually prefer to build something NEW!!! Than fight with other people's bugs and screw ups ANY day!!! At least you get a chance to start afresh, with all the latest toys and whizz bangs, and hopefully do it properly this time around, given what you already know! Well... That's why we use versioning in our software right?! Because we can always find new ways to upgrade and / or rebuild it! Rather than flog an already dead horse!!!
@@sable4539 Huh? In my personal code and my day job. I've built lots of systems from scratch! And the best way to learn IS to build something new from scratch! I remember transitioning to Full Stack, when a company wanted me to build them a fully fledged, Content Management System from scratch! When I was just a Backend Developer! So I had to learn a LOT of Frontend concepts, as I was developing the system. But it was a Great experience! It's hard to learn stuff, by tinkering and trying to figure out somebody else's code... In order to be a Good Developer... You HAVE to get your hands dirty... And develop! Just like being a good pianist requires that you have to go out there and PLAY the piano! Coding is a doing skill... Just as much as it is a thinking skill.
@@DarthVader11912well... How are you going to determine what options there are for the drop down? Presumably it might be different for different people, so now you have to hook into the user state. Then you have to write the logic that determines what options are available for the user. But you also need to determine what the valid options are since you can't hardcode it. So you have to find a way to guarantee parity between the options that are shown and valid options that are recognized by the rest of the application. Then God forbid you have other input that depends on the currently selected option. Or different form validation based on the current selected option. I mean the amount of things surrounding it could be listed endlessly...
4:45 this whole oddly specific rant about Bill and the co-founder's daughter not being able to post cat pictures and three weeks waiting to add a button... I think @ThePrimeTime actually experienced this, and is still recovering. Why is Bill like this?
6:13 JSON with Dr's Handwriting being JSON without type safety... I used to get STRING values for "false"... Let that sink in. if ("false") { console.log("oops"); }
17:53 The End of the World by AlbinoBlackSheep mentioned! I thought I was the only one to still remember that video, absolutely was dying that you quoted it, such classic TH-cam!
19:39 Trailing slash is usually meant to distinguish between directory location and file location. For HTML pages it should be with slash because it is actually loading /index.html. But since it looks ugly this semantic standard is rarely respected.
Thanks for this. Got stressed hard when tried to add an "close on submit" functonality to a really bloated form and thought "dude why can't I get this to work it's just HTML, SQL and JS". Now I know I'm not actually alone.
16:05 A company I worked for had a pretty interesting idea to combat this: Charge the individual departments for IT support and the hardware. So the IT cost of each department can be traced internally and for what the cost is used (which software etc.).
@antoineleduc7611 It sounds like the departments would be forced to make things work so that their budgets/paychecks wouldn't be eaten up by IT expenses. That sounds effective. But, you would have to pay everyone above market rate so they would be working towards a bonus, not just what they're worth.
The funny and counter-intuitive thing about being in a cost center is that management often seem to have no qualms about wasting money hand over fist on pointless cargo cult practices and gratuitous AWS resources and proprietary software licenses. It just means they view you as trash, they don't actually care about the cost in cost center.
Fronted development gets much easier once you stop using endless amounts of libraries to solve every tiny problem - especially those that say solve state management or bring CSS into JS.
This is why I'm so glad I learned RxJS and Angular. I never encountered the need for a state management library, and I was able to have my template rendering be the trigger for API requests all while having sensible caching, and zero misalignment of source data across components.
Sigh. This is so true. And now I'm going to be doing a project where we take ALL our apps and combine them into a single amalgamation equivalent to a mutated Beholder.
My all time favorite is the "Email" Button or link. It usually a last minute add on or has a story description of email data. In the reality what they meant was email the current state of the page with all the filters applied. Later they realize what they really want is to export data to Excel.
They: It's just a button, why is it taking so long? Me: The thing is, this button component is being used on 12 different pages and being used there differently with different props, so because I changed some button's state logic, I need to test it on all 12 pages. BUT THAT'S NOT ALL. The project has different versions based on a user's country, so I need to change the ENV variable, do the "npm run dev", which is gonna compile everything in ~2 minutes, because there's tons of components and it's Nuxt 2, and THEN I can test the button logic on this version, then change env variable again to another country, compile it again in 2 minutes, etc etc.
I feel like, when the client is complaining what it is taking so long, it is just a button, you should slap a button that does nothing. Then say that " just the button" part was done all along, however you are still working on making the button actually work.
No because QA will find a way to call your do-nothing button a security vulnerability, not to mention it takes two extra jira tickets to do that, first ticket to request a change to the original requirements, and second ticket describing the new requirements in more detail (Tod from management said, "it does nothing" isn't enough). Bill will add your buttonStoredProcedure (to appease QA) to dev in half the time it would have taken to implement the full functionality, but only because he actually saw your second email the day you sent it and still hasn't responded to the first.
As a full stack developer, I don't know whether to laugh or cry. This is painfully relatable. At least we don't have to deal with IE 6 anymore, which was the bain of my existence in the 2000s
4 หลายเดือนก่อน
3:30 LMAO THIS IS HOW IT FEELS EXACTLY 6:30 that's because Bob from backend asked Tom the genius to write a company-owned JSON Serializer that can also function as an entire Domain language.
15:20 Best experience as a dev I’ve had is when the businesses core product is the software. Doesn’t have to be a software company but if when it’s foundational to the business the focus is on that, not some other product.
Nothing than working on old technology has ever hit closer to home for me. I am 2 years into the industry and still learning. The only big projects I have worked on professionally other than Bug fixes are a C# .Net Framework 4.6.1 Monolithic API to .Net Core 3.1 Microservice Rewrite (Deadline and Technical Debt nightmare) and 3 AngularJS Projects to Angular 17.
I always and watch appreciate the add drops at the end of videos. I use this time to reflect on what I watched and learned, letting them run. Good video but I miss the blue hair, but at least got a reminder it existed. =]
8:58 Recently, in my job I was "lucky" to be able to upgrade the frontend of one of our customer's webapps, from React 16.8 to React 18.2. The good news was that the components where functional components with hooks. But anyway, it was was a very big pain to upgrade. I suffered hundreds of breaking changes and several old npm packages with no compability with React 18.2, forcing me to research new packages and adapt to them. I also had to upgrade the backend from .Net Core 3.1 to .NET 8. But compared with upgrading the frontend, this was a walk of the park. The bad news is that there is another webapp of the same customer that is older and it's full of class components without hooks. We plan to upgrade it in the future. That one will hurt for sure 😢
Upgrading from .Net Core 3.1 to any later versions is a walk in the park. .Net is extremely easy to upgrade, that is why I upgrade to the latest every year. There is almost never an issue.
I once lost an entire day when we upgraded to React 18 because it was incompatible with react-beautiful-dnd. We used react-beautiful-dnd. Extensively. We needed to upgrade. Yet many resigned themselves to the fact that we simply would not be able to. I sifted through all of Github issues/SO/etc until I finally found some janky solution that somehow (I still don’t fully understand how) fixed the issue. We now use React 18.
@@evancombs5159 I agree. I remember some years ago trying to upgrade a .NET Framework 4 app to .Net Core 3.1 and that was ver painfull 😢 But recently I upgraded 3 apps from .Net Core 3.1 to .NET 8 in less than 2 work days 😊 The only big problem we had was no directly related with .NET itself, but with a library used to generate excels called EPPlus, which sadly become a paid product in their latest versions. We tried using the last non-paid version, since uses .NET Standard 2.0. But that didn't work because the library uses System.Drawing, which after .NET 6 only works in Windows, but not in Linux. So, we had to install a completely different library (Open XML) and refactor the generation of almost 20 excel reports 😔
@@evancombs5159 This is like my third attempt to reply you. I don't know why TH-cam doesn't show my comment... Anyway, I agree with you. I remember some years ago I tried to upgrade a .NET Framework 4 app to .Net Core 3.1 and that was very awful. But recently I upgraded 3 apps from .Net Core 3.1 to .NET 8 in less than 2 work days. The only big problem I had was no directly related with .NET itself, but with a library used to generate excels called EPPlus, which sadly become a paid product in their latest versions. We tried using the last non-paid version, since uses .NET Standard 2.0. But that didn't work because it uses the System.Drawing, which after .NET 6 only works in Windows, but not Linux. So, we had to install a completely different library (Open XML) and refactor the generation of almost 20 excel reports...
@@evancombs5159 that's good to hear, I've dabbled in dotnet core a few times but I know they've had a lot of major releases over they years and didn't know if some of those were big breaking changes or not.
We still have s into our MooTools code, Angular 1 (non-d) and stuck on React 17. It either takes a year to major bump a core library or one engineer takes a month or two and does it for everbody and prays the coordinated release somehow actually works.
I remember working at Atlassian and watched a senior dev and intern spend 2 months refactoring a 'Button'. Also at Atlassian I remember needing to go through 3-4 layers of teams to make a 1/4 line CSS change.... haha
I wrote an AJAX function that worked with IE6 and Firefox, this is due to jQuery already being implemented in the Liferay framework but it was a custom version that didn’t have all of the features and it would be more difficult to teach everyone to use different versions of jQuery for different purposes when all we needed was the AJAX functionality. This was a rushed application (built in 6 months) that was partially replaced a few years later, but the billing process was so complex that it couldn’t be easily replaced so a bridge service would take the data from the new one and put it in the old one to complete billing. It was filled with bugs that towards the end of life for the old application, it would crash often due to memory leaks and require restarts, which everyone entering their billable time would then have to wait for the system to come back up. This is why software engineering is hard, you can build anything quickly, but to have a long lived service requires much more effort and thought.
so happy to theprimeagen explaining the button abstaction, I worked at a company and they had abstraction code for https requests, and it was fking confusing how that code worked and how to include payload, query params and whatnot, and it was my first day there and it was taking me a while figuring out what this code does, and my seniors, instead of helping they were like, "why is it taking so long", I'm like this is my first day and this code base is 20k lines, with shitty abstractions. So yeah, I left that company. Glad I did.
i was in that position once. I coded so fast they just LOVED the end result. to this day t he mentions my spaghetti code on his 2 decade old weebshite. hahaha It was just a contacts system, looked amazing, tottally unusable codebase. I couldn't understand it when I finally stopped going. worst job EVER. wooden chair so many f'd up abusiver shlt this goof did .. haha i still chuckle, they were SOOO impressed they nevetr checked my source. ps they were all supposed maguic devs and had HTML tags wrapped in ddiff funbctions . but wrong were it took more work to use. it was wack.. my html wrappers saved a tonne of time and made forms so easy to read/dev/process/update/insert etc
this is why comments are important. Communication is key. Based off this is reaction it's clear to me we've over develop certain apps and programs. I think it's time we simplify the net. Web 4.0 should be a return to form and structure and a standardization of multi-module design. Also, how about up as programming language that's all emoji's and memes
The problem I have with frontend is that when I finish a feature, I actually want to admire my work. That's usually where my problems start. Do the colors match the color scheme of the page? Suddenly, I realized that the button needed to change color when hovered. Oh no, it needs to have a ripple effect when clicked. What other events can happen to this button, and how should it look with each one? What if the request failed? The button needs to be disabled until the form is edited again. And now we're into form validation: should the button even be clickable if any of the form fields are invalid? Sh*t! How do I know what validation rules to use? Should this be done in the front end or the backend? I probably need a new Jira ticket for this. Don't even get me started on accessibility (a11y). Does the button need tooltips? Animations? How do I disable animations if "prefers-reduced-motion". Tailwind of course! Do we even need a button? Can't we use the anchor element styled as a button and handle the click with extra JS, or should I have used a div in the first place? Finally, at the end of the day, it hits me: I need a dedicated React component for this button. And that's how one button took more than a day.
10:35 same for me but with a codebase of angularjs + angular 9 + angular 17 + svelte because one guy loved svelte but he left XD oh and a backend in php and java and kotlin and scala because one guy loved scala but, you guessed it, he also left.
Flash was and still is ahead all languages in the front-end and back-end development. With Flash and Action Scripts you would see in real time what you create and have a product done in days vs years with other languages. Because Flash started using GPU Steve Jobs made a deal with Adobe to kill it. Even in 2024 we still don't have tools to create applications in real time that works with any shape or form. Every language is a nightmare dealing with GUI and back-end code. We need tools like flash did, write you code and in real time you see the final product in both front and back-end. In flash you would write one line of code and other languages pages and pages to create the same thing. Front-end development is a nut job and slow
I'm constantly wishing I was 42 now instead of 22 so I could have caught the simpler old days. Or the opposite, if only now did stuff start growing like this. Idk, graduating in half a year and idk where to start. Learned frontend, backend, android dev, java desktop dev, low level, high level, etc... and I just... so much stuff to learn to get going and I can't find a field of passion.. Sigh. Im taking 6mo after graduation to build a project in each field, hoping that might find me a passion in one of the fields
You don't "find" a passion, you build it yourself gradually by first going through beginner hell, meaning: period of frustration, things being slow, boring or not working, a "valley of despair" as some people call it - and then coming out at the other side of it with skills you've acquired that get you in the flow state ("passion") and are actually useful to the marketplace so you can get the financial rewards associated with that. Also, stop wishing you could have caught "simpler old days" because those "simple" days were not simple at all. For one, we didn't have NEARLY as many learning resources as we do now. Everything had to be figured out first, and it was HARD. The further you go back in time, the less you have books, courses, google, internet and ChatGPT available, up to a point where you have NONE of it. Do you think tech is "easier" in such a world? Think again.
Lol reminds me of that EOD that was running for some 12 hours, never finishing before the bank's opening the morning after, because the CEO wanted everything to go through a crystal reports and you had a crystal box dangling in the architecture bottlenecking everything, and I was the one that figure it our with stroke of genius: "turn that box off, run the EOD again" and it ran in 4 hours... Did I mention their solution was to change the transaction dates to make them look like they ran within the day, for those that know banking apps, it is very funny. Sometimes is less what you know than how you look at things.
The good old days! I remember forgetting I had an interview one day, and thought, F it I'll turn up straight from the beach. I was wearing a singlet, flip flops - we call them thongs here in Australia but I digress - and had no laptop. I blazed through the interview in the most I don't give a f*** about the outcome of this ever. I had no way to log into any of my sites to demo my code because, well I can't remember. Anyways I got the job, and when I started I realized it wasn't even a dev job. It was a designer job. And I killed it. The good old days. That was in the advertising industry where they THREW money at your face. I was once asked if I could 'dev a video ad for the cricket'. Ofcourse I said yes and learned Adobe After Effects overnight and delivered. The GOOD OLD DAYS! haha I am a husk of that former self right now. Riding very close to not affording my rent, but you know that, I'm still learning Go lang haha I love being a dev!
Dude this is so real I am a Frontend dev for a saas startup which provides websites. I am currently working on 3+ projects and get paid like shit. I should be making $100k+ but making $10k+
That whole bit about the button needs to be a separate video so front-end devs can show it to their boss every time they complain about how long it's taking to add a "simple" button.
You don't know pain until you've jumped in a 3 yo react project that's gone through many dif teams, no-ones taken the time to come up with a set of design patterns to adhere too, and the most recent team is mostly bootcamp grads on their first job. For me it was a set of totals that were calculating things incorrectly - one whole feature re-write later... I got the thing to add correctly.
Oh sweet summer child. Try debugging kernel bugs from the time you were in middle school in a massive project started 15 years ago and somehow still not in production yet. Bonus point if most of your colleagues are 55 years old ex-electronic engineers never trained in software engineering and computer sciences.
The same reason I question the current trend to trust the unit tests written before you Jim Coplien is write when he said today motto in Dev is "we don't need to know the system, we just need to improve it" 😂
So far I love that it's getting harder. I think the complacency and unconditional job safety were suffocating the industry and ruining the quality of work. I still suffer from this at my workplace... Kindergarten scrum democracy, catering to the weakest link in the chain, absolutely no accountability, some devs produce virtually no value with a month of work, etc.
9:00 Yep that's me, I thought I would be working on the most cutting technology, but Instead I am refactoring a node.Js v 12.0 and react 16 codes written by people who are long gone with 0 documentaion
I am a junior Frontend developer at a huge Indian firm and we're building a Calendar widget for the client and my lead who doesn't know shit about frontend development keeps on nagging me about "Oh, it's just a calendar you just have to load the data" meanwhile the library used has turned obsolete, there's no support for newer issues and they changed half the shit and didn't even document it!!!
Your description of "JSON in doctors handwriting" sounds exactly like what is usually returned by the rest API of an average wordpress plugin (or probably even wordpress itself)! Pain!
"It's just a button, why is it taking so long?" is a sentence capable of instantly curing low blood-pressure conditions. Only downside is the instant depression.
With the button connected to 20 functions lol.
You frontend developers are chill. Imagine us full-stack developers, who not only have to do frontend of the button + backend api changes, but also kubernetes stuff 🤣
hahahahah TOO REAL!!!!!!!!!!!!!
idk man as a full stack its easier when you have control of both @@przemekh4857
lol that button thing hit home. Client wants a "print" button put on a data table and added to the context menu. Ok so just print the page? No, a custom-tailored report with specific formatting and watermarks. The data they want exists across 18 tables and in 1-1, 1-many, and many-many related relationships. The data fields they want keep evolving, as well as the formatting each time they see the prototype report, as well as the report will display different data fields and be in different orders depending on certain conditions of said data. Is it a Template A? Does it contain X number of software items? Is it Tuesday!?!? 2 months later...."Why is it taking so long, IT'S JUST A PRINT BUTTON!!
i feel your pain
Why are you not paying me? Its just a transfer button! Counter attack
I feel seen
"You think you know what it takes to tell the user it's their birthday? You know nothing of my pain. Of Galactus's pain!"
Sounds like a requirements nightmare. Force them to specify the requirements, maybe they'll realize how crazy it is. At some point, you have to politely say no and make them understand why.
"You just want to add a button but Bill is on vacation" is the realest thing I've ever heard. That has been my entire experience in software, just constantly waiting for [person] to get back from PTO or vacation so that you can do the one thing that's preventing you from doing everything else you need to do.
10 years ago every engineer my work was blocked for a week because the TFS build engineer was out for a week and had exclusively locked a file. We just had to work without source control for the week I think.
@@georgehelyar I wanna know the rest of the story, what happened to that guy?
@@georgehelyar During my last internship, there was a corporate merger and all the interns were completely locked out of the system for the last 10 days of the internship. This happened because they were doing a really ugly layoff technique where they deleted the log-in credentials of everyone who was getting laid off. This included the log-in credentials of every intern and they laid off the guy who was in-charge of everyone's credentials so we just had to sit on our thumbs for the last two weeks of that internship before returning the company laptops. And big tech wonders why everyone is leaving for trade jobs.
That’s the opposite of my experience in software. I do my own thing (solo dev), so if something needs to be done, I just do it. Unless I procrastinate.
Pitifuls of one person being the SME.
I'm mainly a backend C# and SQL developer but can do frontend as well. One mid size company I worked with sales to put a "Tax Exempt" checkbox on a page. This was later in the day. Next morning she asked if the checkbox was done!!!! I said no... she said it's just a checkbox! I asked if she just wanted the checkbox there or does she want it to work also. So had to change 50+ C# files in backend code,, some reports, and some stored procedures to get it to work. Guess she thought putting the checkbox on there that the application would know what to do with it? LOL
Yeah... It's not just front end! As Full Stack, I feel like I'm getting disillusioned at ALL ends! Front AND back! Because the backend is fast catching up in being caught up in unnecessary hyper-complexity too!!! With Everything being distributed into tiny little pieces all over the cloud! With each little piece being an entire LAW unto itself!!!
Exactly she couldn't conceptualize the logic behind the scenes.....this is why people hate dealing with clients directly lol.
"You want it there or you want it to work". I'm using that from now on lol
The startup I worked at: "WE WANT QUALITY! MAKE QUALITY STUFF!" (They didn't tell us that he wanted a pretty rapid prototype, so I wrote software to be decently stable, and the CEO hated that).
Surprisingly, the "Jira ticket for that" event almost happened when some guy in the building went to a local grocery store to buy a bunch of grapes for a snack. Middle manager was flipping the you-know-what out. "Bruh. He's gone for maybe 15-20 mins. If that's enough to derail the project, this company is screwed".
Grapes > capitalism
If you are a startup or a consultant, getting the thing working quickly is the only acceptable path. Write as high-quality code that you can, but don't fret about making mistakes, and if someone harasses you about it, ask them if they want something that works or is extremely high-quality and feature-rich. Pick one.
@@Joe-ku1ko True, but we are talking about a guy that can't actually say it straight out. Turns out what he wanted, with us trying to psychically read his mind, was just some UI to show off. The thing didn't even have to work! Younger me couldn't figure that out, but he could have just said it directly.
Then again, if you did this, "Why isn't our stuff stable?" I actually told him that you can't expect something like that to actually work without crashing a lot, and he did not like that. You know what Prime says about 10Xers? It applied at that place. The perceived 10Xer didn't make stable stuff, but the CEO got what he wanted
@@Joe-ku1koas someone working at a consultant firm, making in towk is king. Everything else is a nice to have (except documentation, we are very careful with having everything documented)
@@Joe-ku1ko👍 The age old question - quality vs. quantity.
I cried this week. Something that has been working for years was breaking in very annoying ways that could not be replicated in debug mode and the failures were the exact failures I was trying to avoid in my design. Then, I took a moment to refactor a stupidly written class (I don’t know what the was I thinking. Then again, I’m support, IT, sys engineer, project management, devops, tech lead, and lead software engineer so maybe I forgot a hat when I wrote it). The class affected my task scheduler which indirectly affected the lifetime of triggered tasks which affected whether some state was committed to disk as expected. Since I had to keep prod flowing while figuring out wth was happening, I cried every day…
Software dev is truly an exercise in not crying.
Big blast radius = bad design
Cry early and often ❤
Skill issue
Stay strong, my friend! It's just a job.
Maybe I am crazy or entitled, but aint no way I am doing work for entire company. Should you also be a chef for your company?
The “button diagram” was exactly how my instructor explained docker to me
4:40 as a famous developer once said:
"You silly product manager! You know nothing about GALACTUS's pain, my pain.
Delivering this feature goes against everything I know to be right and true, and I will sooner lay you into this barren Earth than entertain your folly for a moment longer."
I love that video with passion
What video that @Slashx92
4 minutes made into 20. Good job!
5x engineer 🤓
That's the thing I hate about reaction videos in general. I could watch the source and save much time, provided most reactions would won't add much in general 🤷
at least he's adding actual content, not like ppl who just watch the video and call it a "reaction"
I watch him at 5x, problem solved
Then shouldn't you just watch the original videos if the reactions aren't what you want anyway? @@gradientO
That "net is built with old tech" part hits hard. I recently had to review a 10 years old app that somehow got a new version, but security team did not want to approve it. I checked the code: Webpack 1.0, libraries so old you can't upgrade them without upgrading / replacing dozens more, tons of libraries on top of tons of other libraries including an internal library for components that was acquired from one of the child companies and was no longer maintained. And the problem? Dep of a dep of a dep 5+ levels deep should be upgraded because string method X is bad or some minuscule shit like that. Which would require upgrading the main dep, which would require other deps changes, yadda yadda. After a day of trying to not colapse the Jenga tower I was told to write down why this can't be done, security team approved another of very many exceptions, and I went back to my oryginal project.
I felt the MTB rant. Currently am in the middle of the same thing happening in our project. It started as a simple enough web application, a .NET GUI and an Oracle DB, the DB had what passed as the application layer in PL/SQL. One webservice of some large data inputs we got once a week, that the customers then could work with on the GUI, and a DB Link for a project that ours spawned off from.
Then came middle management and wanted scalability.. for an app that has maybe 50 users, country wide, max. So now we have an API between GUI and DB, and spend half a year ripping the app layer from the DB, so it can run on multiple servers, if needed (Not needed yet and probably never will be).
We since added a ticket system that needs to communicate to the sister project, so got 2 APIs for that too, one for our end, and one for their end. We also needed a job service to fetch data from a third party data source, actually make that two for another one, because microservices, amiright fellas?
I have 6 different .NET solutions in 3 different versions ranging from Framework 4.8 to .NET 8. Sometimes I wonder where it all went wrong.
This happens with any project.
The bigger the product, the bigger the binary decision tree and the longer it takes to add even the simplest of features.
Eventually, it's cheaper to make a new project than it would take to fix all the bugs in the old project.
Lol yeah but it definitely is worse in some orgs than others, I exited my startup and went to work pretty high up at an established PE firm. They had an internal platform that was ANCIENT; I was assigned to getting it up to date. The engineers there were INSISTENT that it would take months to introduce just some pretty basic ideas I had to make the UX better. They claimed it couldn’t be done. So over one weekend I rewrote the entire front end in React and ported over the backend to a Node MVC microservice. Showed it to them on Monday and said this is what they’d be working with. The team lead was batshit and started yelling at me, saying there was no way they could all learn the new stack and he wasn’t going to do it. So I fired him. He said there was no way they could get this project done without him. Surprise, surprise… two weeks later all of the features and more that I wanted were implemented and deployed to users. Not only that, but the devs were much happier. Eventually promoted one who was really eager to learn to lead, and he ended up taking one of my roles when I left. Think he’s still there.
I actually prefer to build something NEW!!! Than fight with other people's bugs and screw ups ANY day!!! At least you get a chance to start afresh, with all the latest toys and whizz bangs, and hopefully do it properly this time around, given what you already know!
Well... That's why we use versioning in our software right?! Because we can always find new ways to upgrade and / or rebuild it! Rather than flog an already dead horse!!!
@@sdwoneHow did you learn to build new things by yourself?
@@sable4539 Huh? In my personal code and my day job. I've built lots of systems from scratch! And the best way to learn IS to build something new from scratch!
I remember transitioning to Full Stack, when a company wanted me to build them a fully fledged, Content Management System from scratch! When I was just a Backend Developer! So I had to learn a LOT of Frontend concepts, as I was developing the system. But it was a Great experience!
It's hard to learn stuff, by tinkering and trying to figure out somebody else's code... In order to be a Good Developer... You HAVE to get your hands dirty... And develop! Just like being a good pianist requires that you have to go out there and PLAY the piano!
Coding is a doing skill... Just as much as it is a thinking skill.
One time I was assigned to change a button to a drop-down with a few options. It only took 2 weeks and 13 file changes to make it happen.
oh my that is sad
But why?
It's literally just a few lines of js wtf
@@DarthVader11912assuming good structure of the codebase, but never underestimate the creativity of people to write horrendous codebases
@@DarthVader11912well...
How are you going to determine what options there are for the drop down? Presumably it might be different for different people, so now you have to hook into the user state. Then you have to write the logic that determines what options are available for the user. But you also need to determine what the valid options are since you can't hardcode it. So you have to find a way to guarantee parity between the options that are shown and valid options that are recognized by the rest of the application.
Then God forbid you have other input that depends on the currently selected option. Or different form validation based on the current selected option.
I mean the amount of things surrounding it could be listed endlessly...
As my company's Bill, I only hear tiny violins
Man that explanation at the beginning. That went from nirvana to my current work really fast
I'm in shambles after watching this. You took all hope out of my life.
4:45 this whole oddly specific rant about Bill and the co-founder's daughter not being able to post cat pictures and three weeks waiting to add a button... I think @ThePrimeTime actually experienced this, and is still recovering. Why is Bill like this?
4:49 - I can swear I heard Michael Scott right there.
6:13 JSON with Dr's Handwriting being JSON without type safety... I used to get STRING values for "false"... Let that sink in.
if ("false") { console.log("oops"); }
I think this is my second favorite video of yours. Right up there next to PayPal's pASSWORD.
Same here
Loled at least twice and sorrowed once (because of the “this is the dev’s life” remark).
17:53 The End of the World by AlbinoBlackSheep mentioned! I thought I was the only one to still remember that video, absolutely was dying that you quoted it, such classic TH-cam!
It made me unreasonably happy to hear somebody else reference that animation.
return JSON in doctors handwriting KIIILLLLED MEEEEE
19:39 Trailing slash is usually meant to distinguish between directory location and file location. For HTML pages it should be with slash because it is actually loading /index.html. But since it looks ugly this semantic standard is rarely respected.
That was true in the old days where a url mapped to a file in the file system. It's not been like that for decades now.
@@radadadadeei mean url are still mapped to files, just indirectly and with gateway in front that are proxying requests in a number of way
@@antoineleduc7611 if by "files" you mean files on disk, then it's definitely not true. If by files you mean an HTML response on memory, then maybe.
Thanks for this. Got stressed hard when tried to add an "close on submit" functonality to a really bloated form and thought "dude why can't I get this to work it's just HTML, SQL and JS". Now I know I'm not actually alone.
16:05 A company I worked for had a pretty interesting idea to combat this:
Charge the individual departments for IT support and the hardware.
So the IT cost of each department can be traced internally and for what the cost is used (which software etc.).
This sounds like a terrible idea no ?
Wouldn’t everyone just try to make things cheap then
Even when cost would be ideal for long term product quality
@@antoineleduc7611 I am not sure what you mean with "make things cheap"
@antoineleduc7611 It sounds like the departments would be forced to make things work so that their budgets/paychecks wouldn't be eaten up by IT expenses.
That sounds effective. But, you would have to pay everyone above market rate so they would be working towards a bonus, not just what they're worth.
@@iMagUdspEllr well, the same company has a profit sharing policy for the employees (every employee, including the cleaning personell)
FUUUUCK SOMETHING DEEP INSIDE ME HAS BEEN TICKKELED @13:57 IT HITS DEEP
omfg you diagraming the “button problem” was truly beautiful - I’m gonna save this video and require all account managers to watch this
The funny and counter-intuitive thing about being in a cost center is that management often seem to have no qualms about wasting money hand over fist on pointless cargo cult practices and gratuitous AWS resources and proprietary software licenses. It just means they view you as trash, they don't actually care about the cost in cost center.
Fronted development gets much easier once you stop using endless amounts of libraries to solve every tiny problem - especially those that say solve state management or bring CSS into JS.
This is why I'm so glad I learned RxJS and Angular. I never encountered the need for a state management library, and I was able to have my template rendering be the trigger for API requests all while having sensible caching, and zero misalignment of source data across components.
Bro, you're such a vibe, I appreciate you
Sigh. This is so true. And now I'm going to be doing a project where we take ALL our apps and combine them into a single amalgamation equivalent to a mutated Beholder.
As someone who started doing front end in 2012 and then changed career paths…good to see my skills are still top of the line 😂
My all time favorite is the "Email" Button or link. It usually a last minute add on or has a story description of email data. In the reality what they meant was email the current state of the page with all the filters applied. Later they realize what they really want is to export data to Excel.
They: It's just a button, why is it taking so long?
Me: The thing is, this button component is being used on 12 different pages and being used there differently with different props, so because I changed some button's state logic, I need to test it on all 12 pages. BUT THAT'S NOT ALL. The project has different versions based on a user's country, so I need to change the ENV variable, do the "npm run dev", which is gonna compile everything in ~2 minutes, because there's tons of components and it's Nuxt 2, and THEN I can test the button logic on this version, then change env variable again to another country, compile it again in 2 minutes, etc etc.
feels like you somehow should be able to test using code
You can use any tech stack you like, as long as it's .NET, SQL Server, and Bootstrap.
😂😂 dude, you fucking nailed this one
I feel like, when the client is complaining what it is taking so long, it is just a button, you should slap a button that does nothing. Then say that " just the button" part was done all along, however you are still working on making the button actually work.
No because QA will find a way to call your do-nothing button a security vulnerability, not to mention it takes two extra jira tickets to do that, first ticket to request a change to the original requirements, and second ticket describing the new requirements in more detail (Tod from management said, "it does nothing" isn't enough). Bill will add your buttonStoredProcedure (to appease QA) to dev in half the time it would have taken to implement the full functionality, but only because he actually saw your second email the day you sent it and still hasn't responded to the first.
As a full stack developer, I don't know whether to laugh or cry. This is painfully relatable.
At least we don't have to deal with IE 6 anymore, which was the bain of my existence in the 2000s
3:30 LMAO THIS IS HOW IT FEELS EXACTLY
6:30 that's because Bob from backend asked Tom the genius to write a company-owned JSON Serializer that can also function as an entire Domain language.
15:20 Best experience as a dev I’ve had is when the businesses core product is the software. Doesn’t have to be a software company but if when it’s foundational to the business the focus is on that, not some other product.
Nothing than working on old technology has ever hit closer to home for me.
I am 2 years into the industry and still learning.
The only big projects I have worked on professionally other than Bug fixes are a C# .Net Framework 4.6.1 Monolithic API to .Net Core 3.1 Microservice Rewrite (Deadline and Technical Debt nightmare)
and 3 AngularJS Projects to Angular 17.
I work in enterprise and I'm literally writing a wrapper for a bunch of AJAX calls on new projects because that's the approved way of doing things
I always and watch appreciate the add drops at the end of videos. I use this time to reflect on what I watched and learned, letting them run. Good video but I miss the blue hair, but at least got a reminder it existed. =]
LOL "all you want to do is create a button! but BILL IS ON VACATION!"
I just peed myself on "Bill's on Fking vacation" 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣
8:58 Recently, in my job I was "lucky" to be able to upgrade the frontend of one of our customer's webapps, from React 16.8 to React 18.2.
The good news was that the components where functional components with hooks. But anyway, it was was a very big pain to upgrade. I suffered hundreds of breaking changes and several old npm packages with no compability with React 18.2, forcing me to research new packages and adapt to them.
I also had to upgrade the backend from .Net Core 3.1 to .NET 8. But compared with upgrading the frontend, this was a walk of the park.
The bad news is that there is another webapp of the same customer that is older and it's full of class components without hooks. We plan to upgrade it in the future. That one will hurt for sure
😢
Upgrading from .Net Core 3.1 to any later versions is a walk in the park. .Net is extremely easy to upgrade, that is why I upgrade to the latest every year. There is almost never an issue.
I once lost an entire day when we upgraded to React 18 because it was incompatible with react-beautiful-dnd.
We used react-beautiful-dnd. Extensively.
We needed to upgrade. Yet many resigned themselves to the fact that we simply would not be able to.
I sifted through all of Github issues/SO/etc until I finally found some janky solution that somehow (I still don’t fully understand how) fixed the issue.
We now use React 18.
@@evancombs5159 I agree.
I remember some years ago trying to upgrade a .NET Framework 4 app to .Net Core 3.1 and that was ver painfull 😢
But recently I upgraded 3 apps from .Net Core 3.1 to .NET 8 in less than 2 work days 😊
The only big problem we had was no directly related with .NET itself, but with a library used to generate excels called EPPlus, which sadly become a paid product in their latest versions. We tried using the last non-paid version, since uses .NET Standard 2.0. But that didn't work because the library uses System.Drawing, which after .NET 6 only works in Windows, but not in Linux. So, we had to install a completely different library (Open XML) and refactor the generation of almost 20 excel reports 😔
@@evancombs5159 This is like my third attempt to reply you. I don't know why TH-cam doesn't show my comment...
Anyway, I agree with you.
I remember some years ago I tried to upgrade a .NET Framework 4 app to .Net Core 3.1 and that was very awful.
But recently I upgraded 3 apps from .Net Core 3.1 to .NET 8 in less than 2 work days.
The only big problem I had was no directly related with .NET itself, but with a library used to generate excels called EPPlus, which sadly become a paid product in their latest versions. We tried using the last non-paid version, since uses .NET Standard 2.0. But that didn't work because it uses the System.Drawing, which after .NET 6 only works in Windows, but not Linux. So, we had to install a completely different library (Open XML) and refactor the generation of almost 20 excel reports...
@@evancombs5159 that's good to hear, I've dabbled in dotnet core a few times but I know they've had a lot of major releases over they years and didn't know if some of those were big breaking changes or not.
We still have s into our MooTools code, Angular 1 (non-d) and stuck on React 17. It either takes a year to major bump a core library or one engineer takes a month or two and does it for everbody and prays the coordinated release somehow actually works.
1:45 This is why it's a 20m+ video...Love that he goes off on this tangent, btw. Totally worth the 2 minute deviation.
Omg, hearing you use the cockles line brings back so many memories of Webfilings. Don't forget about the sub-cockles.
I remember working at Atlassian and watched a senior dev and intern spend 2 months refactoring a 'Button'. Also at Atlassian I remember needing to go through 3-4 layers of teams to make a 1/4 line CSS change.... haha
For a while I though he was saying "butt requests". I had to turn on captions and then I realised he was actually saying "button requests".
😭😭😭 the original video and your comments, 100% accurate and I hate my life.
I wrote an AJAX function that worked with IE6 and Firefox, this is due to jQuery already being implemented in the Liferay framework but it was a custom version that didn’t have all of the features and it would be more difficult to teach everyone to use different versions of jQuery for different purposes when all we needed was the AJAX functionality. This was a rushed application (built in 6 months) that was partially replaced a few years later, but the billing process was so complex that it couldn’t be easily replaced so a bridge service would take the data from the new one and put it in the old one to complete billing. It was filled with bugs that towards the end of life for the old application, it would crash often due to memory leaks and require restarts, which everyone entering their billable time would then have to wait for the system to come back up.
This is why software engineering is hard, you can build anything quickly, but to have a long lived service requires much more effort and thought.
4:45 Enjoy your stay in Hell - *Bill*
Your Here Forever
so happy to theprimeagen explaining the button abstaction, I worked at a company and they had abstraction code for https requests, and it was fking confusing how that code worked and how to include payload, query params and whatnot, and it was my first day there and it was taking me a while figuring out what this code does, and my seniors, instead of helping they were like, "why is it taking so long", I'm like this is my first day and this code base is 20k lines, with shitty abstractions. So yeah, I left that company. Glad I did.
i was in that position once. I coded so fast they just LOVED the end result. to this day t he mentions my spaghetti code on his 2 decade old weebshite. hahaha It was just a contacts system, looked amazing, tottally unusable codebase. I couldn't understand it when I finally stopped going. worst job EVER. wooden chair so many f'd up abusiver shlt this goof did .. haha i still chuckle, they were SOOO impressed they nevetr checked my source. ps they were all supposed maguic devs and had HTML tags wrapped in ddiff funbctions . but wrong were it took more work to use. it was wack.. my html wrappers saved a tonne of time and made forms so easy to read/dev/process/update/insert etc
this is why comments are important. Communication is key. Based off this is reaction it's clear to me we've over develop certain apps and programs. I think it's time we simplify the net. Web 4.0 should be a return to form and structure and a standardization of multi-module design. Also, how about up as programming language that's all emoji's and memes
@6:57
thanks for the pause, saved me the trouble while I also burst into laughter.
This is one of the best videos! Well done!
4:39 Are you okay, bro? Sounds like you just had some PTSD 😂
"Bill is on Find VACATION!" oh my god.... too real man
The problem I have with frontend is that when I finish a feature, I actually want to admire my work. That's usually where my problems start.
Do the colors match the color scheme of the page? Suddenly, I realized that the button needed to change color when hovered.
Oh no, it needs to have a ripple effect when clicked. What other events can happen to this button, and how should it look with each one?
What if the request failed? The button needs to be disabled until the form is edited again.
And now we're into form validation: should the button even be clickable if any of the form fields are invalid?
Sh*t! How do I know what validation rules to use? Should this be done in the front end or the backend? I probably need a new Jira ticket for this.
Don't even get me started on accessibility (a11y). Does the button need tooltips? Animations? How do I disable animations if "prefers-reduced-motion". Tailwind of course! Do we even need a button? Can't we use the anchor element styled as a button and handle the click with extra JS, or should I have used a div in the first place?
Finally, at the end of the day, it hits me: I need a dedicated React component for this button.
And that's how one button took more than a day.
man, prime unhinged in the middle of the video. this might be a real experience lol
10:35 same for me but with a codebase of angularjs + angular 9 + angular 17 + svelte because one guy loved svelte but he left XD oh and a backend in php and java and kotlin and scala because one guy loved scala but, you guessed it, he also left.
"Touch of Svelte somewhere in there...", I don't appreciate being called out this way.
I got hired by start ups before they were called start ups and we used php and javascript like real men
Me, currently upgrading dependencies in a docker image: "Thank goodness I'm not a front end dev."
Flash was and still is ahead all languages in the front-end and back-end development. With Flash and Action Scripts you would see in real time what you create and have a product done in days vs years with other languages. Because Flash started using GPU Steve Jobs made a deal with Adobe to kill it. Even in 2024 we still don't have tools to create applications in real time that works with any shape or form. Every language is a nightmare dealing with GUI and back-end code. We need tools like flash did, write you code and in real time you see the final product in both front and back-end. In flash you would write one line of code and other languages pages and pages to create the same thing. Front-end development is a nut job and slow
Perfect ending at 5:09
Not even 10 seconds in and they're already coming for you in this video
The Pain Of TH-cam: The stock footage of sipping on a latte is sipping on an espresso.
"It stopped at React 16 because they couldn't do the update to React 18" I felt this in my soul. I'm literally this guy.
Prime, come to Brazil. This is an order!!!
Brazil vs India... Um. Not india 😂
Omg delimited strings representing whatever data is the curse of my company's codebase. Strings everywhere.
I'm constantly wishing I was 42 now instead of 22 so I could have caught the simpler old days.
Or the opposite, if only now did stuff start growing like this.
Idk, graduating in half a year and idk where to start.
Learned frontend, backend, android dev, java desktop dev, low level, high level, etc... and I just... so much stuff to learn to get going and I can't find a field of passion..
Sigh. Im taking 6mo after graduation to build a project in each field, hoping that might find me a passion in one of the fields
devtooling because you get to nerd out about cs
You don't "find" a passion, you build it yourself gradually by first going through beginner hell, meaning: period of frustration, things being slow, boring or not working, a "valley of despair" as some people call it - and then coming out at the other side of it with skills you've acquired that get you in the flow state ("passion") and are actually useful to the marketplace so you can get the financial rewards associated with that.
Also, stop wishing you could have caught "simpler old days" because those "simple" days were not simple at all. For one, we didn't have NEARLY as many learning resources as we do now. Everything had to be figured out first, and it was HARD. The further you go back in time, the less you have books, courses, google, internet and ChatGPT available, up to a point where you have NONE of it. Do you think tech is "easier" in such a world? Think again.
Lol reminds me of that EOD that was running for some 12 hours, never finishing before the bank's opening the morning after, because the CEO wanted everything to go through a crystal reports and you had a crystal box dangling in the architecture bottlenecking everything, and I was the one that figure it our with stroke of genius: "turn that box off, run the EOD again" and it ran in 4 hours...
Did I mention their solution was to change the transaction dates to make them look like they ran within the day, for those that know banking apps, it is very funny.
Sometimes is less what you know than how you look at things.
The good old days! I remember forgetting I had an interview one day, and thought, F it I'll turn up straight from the beach. I was wearing a singlet, flip flops - we call them thongs here in Australia but I digress - and had no laptop. I blazed through the interview in the most I don't give a f*** about the outcome of this ever. I had no way to log into any of my sites to demo my code because, well I can't remember. Anyways I got the job, and when I started I realized it wasn't even a dev job. It was a designer job. And I killed it.
The good old days. That was in the advertising industry where they THREW money at your face. I was once asked if I could 'dev a video ad for the cricket'. Ofcourse I said yes and learned Adobe After Effects overnight and delivered. The GOOD OLD DAYS! haha
I am a husk of that former self right now. Riding very close to not affording my rent, but you know that, I'm still learning Go lang haha
I love being a dev!
oh shoot I thought this was a fireship video the whole time
You know an ad is well placed when prime doesn't even bother to skip it.
Prime totally missed that DT thumbnail (I almost did as well xD)
For some reason when Mr. Prime is mad, he sounds like Micheal Scott from the office played by Steve Carell
Dude this is so real I am a Frontend dev for a saas startup which provides websites. I am currently working on 3+ projects and get paid like shit.
I should be making $100k+ but making $10k+
That whole bit about the button needs to be a separate video so front-end devs can show it to their boss every time they complain about how long it's taking to add a "simple" button.
You don't know pain until you've jumped in a 3 yo react project that's gone through many dif teams, no-ones taken the time to come up with a set of design patterns to adhere too, and the most recent team is mostly bootcamp grads on their first job.
For me it was a set of totals that were calculating things incorrectly - one whole feature re-write later... I got the thing to add correctly.
Oh sweet summer child. Try debugging kernel bugs from the time you were in middle school in a massive project started 15 years ago and somehow still not in production yet. Bonus point if most of your colleagues are 55 years old ex-electronic engineers never trained in software engineering and computer sciences.
@@kakwa”I suffered more than you.”
@@kakwa "Not in production" is where you lost me. IMO things just START to get real at production.
Bonus points for investors with realllly bad ideas.
To date, since 2009, I have never worked in green field project (expect for personal project) at a company. Always legacy
Men, This makes me go crazy goooooo
The same reason I question the current trend to trust the unit tests written before you
Jim Coplien is write when he said today motto in Dev is "we don't need to know the system, we just need to improve it" 😂
TFW you’re looking for a 10x programmer because your codebase is a 0.1x multiplier
So far I love that it's getting harder. I think the complacency and unconditional job safety were suffocating the industry and ruining the quality of work. I still suffer from this at my workplace... Kindergarten scrum democracy, catering to the weakest link in the chain, absolutely no accountability, some devs produce virtually no value with a month of work, etc.
Respect for showing his ad url
I see the Pirate Software inspired paint drawings!
Nice hahaha. Great video!!
i don't think its pirate software inspired, prime has been doing these gimp drawings for over a year or so.
You're new here aren't you?
pirate software just uploaded a short about the etch a sketch feature on mspaint
PS fan but this ain't it chief.
> Prime taking a breath
> "I see the Pirate Software inspired breathing!"
9:00 Yep that's me, I thought I would be working on the most cutting technology, but Instead I am refactoring a node.Js v 12.0 and react 16 codes written by people who are long gone with 0 documentaion
I fucking HATE when a client says "It's just X, why is it taking so long?". It makes me so angry I could punch a baby.
I choose angular, I'm feeling like jesus walking with his cross in the back.
🤣🤣🤣 "A touch of svelte in there because somebody wanted to try it out." 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣
I am literally going through JSON in doctor’s handwriting
Hot damn, I primarily don't do frontend but I felt the paaaaiiiiiiinnnnnnn. 💀
I am a junior Frontend developer at a huge Indian firm and we're building a Calendar widget for the client and my lead who doesn't know shit about frontend development keeps on nagging me about "Oh, it's just a calendar you just have to load the data" meanwhile the library used has turned obsolete, there's no support for newer issues and they changed half the shit and didn't even document it!!!
watching a reaction of a video i already...goodness what are we doing with our lives
Your description of "JSON in doctors handwriting" sounds exactly like what is usually returned by the rest API of an average wordpress plugin (or probably even wordpress itself)! Pain!