Forem: coco

Lo viejo funciona, Juan! A monolithic is enough, almost ever.

coco — Sat, 20 Dec 2025 21:41:40 +0000

It is a common saying: ‘You will need a MVP ready as fast as you can.’
Because you need to test it, to know if there are any users who validates your idea and see if it solves anyone’s problem. However, as a developers, we are always designing a huge architecture for no more than 100 actual users.

I need a landing page with a contact form to input their email and user’s name.

Great!

npm create vite@latest

I will develop a ReactJS project with router, so any case the user writes the URL wrong, I can redirect them to the homepage again. I will split the code into many components. One for the header, the other is the footer, a Main-content wrapping the different sections.
I will need a React Hook Form. For isolating component re-renders, to improve the site’s performance, of course.
To style all this, I will use Tailwind and @gsap/react for animations, so I will be able to finish it faster.

Then, I also need an API in NodeJS to send the email.

A POST route for the form,
GET route for checking the API’s health.
GET default route for index.

I was thinking about using express-xss-sanitizer middleware for security reasons, and Jest for unit testing.
You will have it in two or three weeks.

But we are forgetting the KISS. Where is the love?

KISS rule is Keep it simple s…

Instead of all these files, routes, packages, and repositories. What if we develop a simple HTML with CSS. You can use Tailwind, and GSAP, loading with a CDN if it makes the development faster. But you truly don’t need React for a simple landing page.
You could even try Astro first, or just Vite. But I do believe that a simple HTML file is enough.
You will need a backend to send the email; it is mandatory. In this case, you could look for premade tools to send emails, or serverless such as Lambda in AWS. However, you can do it with a simple PHP file. If you don’t know PHP language, you could ask ChatGPT or GitHub Copilot to make one for you.
With two simple files (one HTML another in PHP), plus a CSS file, you will have the landing running.

Is it the old way to do it? Yes, but it is easier and faster. You will have time to make your micro-service architecture any time later.

The main goal of a developer is to fix any problem as fast as possible. If you can make it cheaper, better :D

You can also do it with no-code apps like Wix, Webflow, Carrd, etc. I know that. However, this post is for developers, not for entrepreneurs. If you are not a developer, you should definitely go to no-code apps. They will make your life easier.

However, what I am trying to debate here, is that we (developers) always think about the first rule: DRY (Don’t Repeat Yourself), but we always forget about the second KISS rule. It is important to balance both.

PHP is going to end! Again!

coco — Fri, 26 Sep 2025 12:02:29 +0000

How many times have I read this? Too many. But PHP is still here, after 30 years. In 2025 the 8.4 version was released, and most of the www was built in the PHP language. So, this will be the php-is-going-to-end 10.0.

I really love coding in PHP. I learnt it before JavaScript, and it saved my life many times. Every time I need to make a quick API, CMS, or E-Commerce, I used to develop it in PHP. I worked with many of its flavors: Laravel, Codeigniter, and -of course- WordPress. I had a huge success, and I am very glad about it. However, this year I decided to quit!

I’m really sad, because I have worked with PHP for more than 10 years. It’s like a friend, and we had to leave it behind.
The issue was that a few years ago, I started having trouble finding junior developers who knew PHP. Educator aren’t including the language in their topics. The results, new developers were coming with the MERN stack. For us, it was a huge problem. We had many projects developed in PHP and I need to make sure someone could support them.

We have many projects running. So, we need to develop an API in NodeJS, and two frontends on ReactJS. I know there are other options, but we took advantage of the refactor and we did it well for once. We need to develop two Frontends (public and private) because the public had a specific design the client approved, and this was harder to develop. While the look and feel of the admin is not important. We used Tailwind or bought a template depending on the case. We decided to make an API was to be fetched from the admin and public together.

All the process had to be transparent, so a regression test was very important. But we also needed to test all the functionalities again. That took a long time as well.

It was such a debate and a hard decision to make. But we needed to convert most of the projects into NodeJS we could, before it was too late. And it was teamwork. Designers looking for old Photoshop files, the account team talking with clients, technical leaders making the plan, and QA working on test cases. The whole process took more than 2 years, but it is almost finished by now.

It’s funny that new technologies couldn’t kill PHP, but now, finally, maybe it’s happening.

Vibe Coding!

coco — Tue, 23 Sep 2025 21:27:39 +0000

I have been Vibe coding. Everybody is calling that shit. Yesterday, I was against the trend; now I jump on the wave. But why? What is it all about? When is it appropriate and when is not?

I have been using GitHub Copilot for over a year, especially since it was free. I tried Cursor, and Gemini is my best friend. However, this is not vibe coding. At least, this is not working well with AI. I was doing it all wrong.

GitHub Copilot is not Vibe coding

This quarter, the CEO sent me on a mission: “I’d like you to test every AI tool you can and make a manual for your team. I want the hole agency working with AI. We must be more efficient.”
Therefore, I pushed myself to test them all for the last three months: GitHub Copilot, Cursor, Claude Code, Firebase Studio, Figma Make, and other tools. I looked up many courses, watched so many videos, and did some experiments. I didn’t want to just test these tools. Hello world was not enough. I wanted to really use them.
That’s why, I made this plan:
First, I made a list of the most important ones and found out what they are for.

Gemini or ChatGPT is the new Google. It is useful to ask questions, investigate, or improve your ideas.
Cursor and GitHub Copilot are developer assistants.
Figma Make, Webflow builder, or Firebase studio are tools for prototyping.
And Claude code just makes the job for you.

Then, I made another list with the issues or tasks the team had in the queue to do. Some were internal jobs, while others were for our clients. I tried to identify which ones could help me the most to test these tools. I picked up four for testing myself and left the others for the team.
Tasks selected:

To upgrade AWS-SDK V2 to V3 and write a new documentation of a 5-years-old API project. (We don’t need this, but I wanted to do it)
To fix one inconsistent bug of an internal tool developed with NodeJS and Handlebars.
A catalog products API demo for a client. As it was a JSON end API, we needed something more illustrative to show our client the functionalities and possibilities of the product. So, we must make an API Frontend and a retailer product page demo.
A landing page with a decision tree logic and a form.

For the first and second task I chose Claude Code. This tool is awesome. You just need to install it, log in, and it does the work for you. Literally.
It is a console tool. So, it can do everything. Since scanning all your project files, till executing console commands. It knows when Docker is running, your NodeJS version, etc. Of course, you could let it do whatever it wants, or you could keep an eye on the tool letting it ask you step by step:

“I will run npm install, yes or no. I will add this package to the project, yes or no. Etc.”

This tool is useful when you are working on an old project, and you need to fix some issues. First, I asked it to read the files and make me a new documentation of the project. After 5 minutes, it wrote me a new .md file with a very detailed documentation. Wonderful! It even mentioned a few features I didn’t remember at all.
Then, I told Claude Code that I need to upgrade the AWS SDK version. So, it started again and after a minute it modified three files:

It added the new package version to package.json, and installed it through a command in the console.
It modified two more files when this service was created and executed.
Then recommended me to test it and make a new .md file with the detailed modifications it did.

It worked fine, of course and took me 30 minutes of work even tested the project online. :D

I don’t know how long it would have taken me to get this kind of work, but I am sure it should have been longer.
The most importantly, this kind of task is not a creative one. You need to upgrade, only because AWS is warning you to do it. You never know exactly why, but you must do it. That’s the main reason to do it with AI: done quickly, efficiently, and focus on something else; more important, perhaps...

The second task was quite different. We had an internal tool with an inconsistent bug. Sometimes it worked but sometimes didn’t. We HAD TO fix it. This was not a warning.

I told Claude Code which was the issue and asked it for help to fix it. It took 1 hour. But finally, it was done. We had to work both this time. Claude Code modified a few files and I tested. Sometimes It fixed something and broke other thing. So, we were iterating changes till It was all fixed. I never had to write a single file. I was just reading the changes, testing the tool, and giving feedback to Claude to continue working.

It took it 1 hour, so I thought it was too long. This time I wanted to compare wether the AI was efficient or not. Therefore, I made a pull request, but I didn’t approve yet. I started a new branch and tried to fix by myself, without AI. However, I couldn’t do it in 1 hour. It was a weird bug. So, I quit and approved the Claude Code job. The AI was ok and helped me with the bug. Now the tool is working, and everybody is using it.

The first two tasks were not vibe coding as I am listening to it. But, I thanked AI for helping with those tasks, because I really didn’t want to complete them. These are the typical kind of stuff that you always leave and nobody will never do. Now, in less than a morning, I completed them. Next!

The demo for a client was more happier task. There wasn’t a template, I had to develop from scratch.
I had asked the PO to make a wireframe of what she wanted. So, I had this wireframe, an API swagger with all the endpoints documented, and a mood-board made by the creative director.
The task was to develop a CRUD for an API had already developed. It was a laptop catalogue. Each product with its properties, images, and texts. I didn’t need to develop the functionalities to create and delete products. However, the Update has more than one functionality. For example: upload images, localizations, delete properties, etc.
Besides, I needed to develop a retailer looked product page frontend. This fetches the data from the API. Therefore, anytime I update the product with the CRUD, the retailer page would show the differences. I had estimated 40 hours at least.

As I had read and watched many videos, I knew I had to start asking Claude to make a prompt for another AI. So, I told Claude what I needed to do and uploaded the wireframe and the mood-board. The LMS, wrote me a long and detailed markdown file. This is very useful because writing these instructions by myself it would take me so long and I would never do it as detailed as the AI. Between bots, they understand each other.
Then I decided to test two similar AI: Firebase for the CRUD frontend and Bolt for the retailer product page. Both worked fine. I used the prompt Claude did for me, the wireframe and mood-board.

For the product page frontend I uploaded a screenshot for a real retailer product page and paste only the specific endpoint that fetch the product with all its properties and assets.
Most of the work was already done. I need to add more CRUD features and localization. So I downloaded the code and run local. I continue working with GitHub Copilot. I started for free, but It didn’t last so I upgraded to pro. The tip here is to go step by step. Example:

Add a button in this page to upload a new image. The button just opens a modal.
Add a form in the modal to upload an image. Drag and drop functionality.
Add a handle form function to get the image file and validate it.
Connect the function to the API and catch the response.
Close the modal and show the new image uploaded in the images’ carousel.

As I experienced, if you just asked it all together, it wouldn’t work fine. It would get confused and would mix it all up. GitHub copilot is a helper; it will not do all the job for you. If you need something like that, Claude Code is better.
This third task took me 20 hours. Half-time I had estimated. It was ok and ready for production. However, I realized that you need to know what you are doing to make it right in the first time. I suspect that a SR developer would get much more out of the tool.

For the last task I had decided to do, I wanted to try something different. I need to develop a landing page with a decision tree logic and a form. The API to send the form was already done, I just need to fetch with the POST method, and that would be all. So, it was only a Frontend job. The difference was that I had a brand guide to follow now. This landing page was a sub-landing of the main site. So, I decided to use Figma Make.
It was excellent. Now we are talking, this is fully vibe coding!

I only needed to tell Figma what I need, and it built a React project with this full landing developed. Then I paste the site URL and ask Figma to copy the design. I downloaded the code and asked Cursor to docker the project. We are using an AWS EKS, so I need the project in a Docker container. Finally, I replaced the text the account team had sent me and published.
4 hours later, the first version was running in production. The best is that the account team (my client) could try it, browse it for a while, test the mobile version and then ask me for changes. This is better than making a Figma prototype. Because it is working, you could make no changes at all and publish. You could try with your audience for better. The best is that it was only 4 hours of work. More agile methodology than this is impossible.

I now understand what Vibe coding is about. It is very useful for prototyping. Having the MVP in the shortest time ever. This is very useful for business, and we need to take advantage of it.

My experience being freelance and developing e-commerce

coco — Wed, 16 Jul 2025 14:53:47 +0000

When I was working as a freelance developer, I developed many e-commerce sites. Most of them were on WordPress using the famous WooCommerce plugin. But I made one with Mercado Shops for a friend. I have never used Shopify till now, and I’m fully converted.
10 years ago, WordPress was the king. It was right because WordPress.org was open source. You could download the code and customize it as you wanted. Of course, you needed to buy a hosting, upload all the code, and install it. Then, you needed a good template and some plugins: WooCommerce (for making WordPress a store in the first place), and a few others for security, contact forms, captchas, payment, etc. You could build it all for free, but you needed to know a little bit or be patient and learnt on your own.

Later, Hostings allows to install WordPress automatically, so you didn’t have to do much work to start your site. However, you will need to install templates, plugins, etc. But the most important thing is that your site need a maintenance. If you had an e-commerce, It would be your tool to earn money. It couldn’t fail because you would be losing sales. WordPress sites, especial e-commerce sites, are being hacked every day. So, the best you can do is to hire someone to secure and back up your site or pray.
Tienda Nube, Mercado Shops, Shopify and these modern tools are different, because they sell you a service. You are buying a hosting, security, and maintenance service, also a backup, per month. As a result, you have a tool to develop your own shop. You can do with no code or coding, but this is equal to WordPress; the main change is that you buy a service. Therefore, you only need to focus on selling, and they will take care of the tool itself.

What does it mean?

It means that you don’t need to worry about security, firewall, network, server, deploys or upgrades. If you work with no code, you don’t need to worry about functionality eighter. It is a turnkey service. You needn’t start from scratch; a simple store could be ready in a day. The simplest one, but it is all you need to start selling.

Nevertheless, if you really need something more advanced, you can still code and develop extra features.
Years ago, when I was THE FREELANCE DEVELOPER, I worked with WordPress. However, today, my advice to any client would be to use Shopify, Tienda Nube or Mercado Shops. Because they have whatever you need to start selling online. You don’t need to reinvent the wheel, do it as simple as you can. When your sales go up, and the commissions you pay are higher than the cost of having an IT department, then you can think of another option.

How to create a beautiful shop in a few clicks

coco — Wed, 09 Jul 2025 15:20:51 +0000

Hey there, aspiring entrepreneurs and future e-commerce moguls! Are you ready to turn your product ideas into a thriving online business? Look no further than Shopify, the leading platform that empowers millions of merchants worldwide. This tutorial will walk you through everything you need to know to set up your very own Shopify store, from the initial signup to launching your first product. Let's dive in!

Why Shopify?

Before we get started, let's quickly cover why Shopify is an excellent choice for your online store:

User-Friendly Interface: No coding required! Shopify's intuitive drag-and-drop interface makes it easy for anyone to create a beautiful store.
Scalability: Whether you're selling a few handmade items or thousands of products, Shopify can grow with your business. Robust Features: From marketing tools and payment processing to shipping solutions and analytics, Shopify has everything you need under one roof.
24/7 Support: Stuck? Shopify's support team is always there to help.

Step 1: Sign Up for Your Shopify Account

First things first, head over to the Shopify website and click on the "Start free trial" button. You'll be asked to enter your email address, create a password, and give your store a name. Don't worry, you can change your store name later if you need to!
You'll also answer a few quick questions about your business, like whether you're already selling products or just starting out. This helps Shopify tailor your experience.

Step 2: Explore Your Shopify Admin Dashboard

Once you've signed up, you'll be greeted by your Shopify admin dashboard. This is your command center, where you'll manage everything related to your store. Take a moment to familiarize yourself with the main sections in the left-hand sidebar:

Home: Your overview of daily activity, tasks, and recommendations.
Orders: Manage all your customer orders.
Products: Add, edit, and organize your products.
Customers: View and manage your customer information.
Analytics: Track your sales, visitors, and other key metrics.
Marketing: Create campaigns, discounts, and automate emails.
Discounts: Set up special offers for your customers.
Online Store: Customize your theme, pages, navigation, and blog.
Apps: Extend your store's functionality with various apps.
Settings: Configure your store's general settings, payments, shipping, and more.

Step 3: Add Your First Product

This is where your store starts to come alive! Go to Products > Add product.
You'll need to fill in the following details:

Title: A clear and concise name for your product.
Description: Provide compelling details about your product, including features, benefits, and specifications. You can use rich text formatting, images, and videos here.
Media: Upload high-quality images and videos of your product. This is crucial for attracting customers! Pricing: Set your price, compare-at price (for sales), and cost per item (for your own tracking).
Inventory: Manage your stock levels, SKUs, and barcodes.
Shipping: Set the weight of your product if you're offering calculated shipping.
Variants: If your product comes in different sizes, colors, or materials, add them here.
Search engine listing preview: Optimize how your product appears in search results.
Product organization: Assign your product to a vendor, product type, collections, and tags to help with organization and filtering.

Click Save product when you're done. Repeat this process for all the products you want to add.

Step 4: Customize Your Online Store's Theme

Your store's appearance is vital for creating a strong brand impression. Go to Online Store > Themes.
Shopify offers a wide range of free and paid themes. You can:

Explore free themes: Shopify provides several well-designed free themes like "Dawn" that are excellent starting points.
Visit the Shopify Theme Store: If you want more advanced features or a unique design, you can purchase a premium theme.

Once you've chosen a theme, click Customize to open the theme editor. Here, you can:

Add and arrange sections: Drag and drop various sections like image banners, text blocks, product carousels, and testimonials to build your homepage.
Edit text and images: Replace placeholder content with your own compelling copy and stunning visuals.
Change colors and fonts: Align your store's design with your brand identity.
Configure header and footer: Customize your navigation menus, contact information, and social media links.
Create essential pages: Go to Online Store > Pages to create pages like "About Us," "Contact Us," "FAQ," and "Refund Policy."

Remember to Save your changes frequently!

Step 5: Set Up Shipping

How will your products get to your customers? Go to Settings > Shipping and delivery.
Here you can:

Create shipping zones: Define where you ship to (e.g., domestic, international) and set different rates for each zone.
Add shipping rates: Offer free shipping, flat rates, or calculated rates based on weight or price.
Set up local delivery or pickup: If you offer these options, configure them here.
Manage packages: Define your default package sizes.

Carefully consider your shipping strategy, as it can significantly impact customer satisfaction and your profitability.

Step 6: Configure Payments

To start accepting orders, you need to set up your payment gateway. Go to Settings > Payments.
Shopify Payments is Shopify's built-in payment processor, offering competitive rates and easy setup. You can also integrate with third-party providers like PayPal, Stripe, and others.

Activate Shopify Payments: Follow the prompts to complete the setup, which usually involves providing some business information.
Add other payment methods: If you want to offer additional options, you can activate them here.

Step 7: Launch Your Store!

You're almost there! Before you go live, double-check everything:

Review your products: Are all descriptions accurate and images high-quality?
Test your checkout process: Place a test order to ensure everything works smoothly.
Check your pages: Are your "About Us," "Contact Us," and policy pages complete?
Verify shipping and tax settings: Make sure they're accurate for your region.

Once you're confident, go to Online Store > Preferences (if your store is password protected) or simply remove the password protection from your store by clicking "Disable password" or "Remove password protection" on your theme customization page, if it's there.
Congratulations! Your Shopify store is now live and ready to welcome its first customers!

What's Next?

Launching your store is just the beginning. To truly succeed, you'll need to focus on:

Marketing and Promotion: Drive traffic to your store through social media, SEO, email marketing, and paid advertising.
Customer Service: Provide excellent support to build trust and loyalty.
Analytics: Regularly review your Shopify analytics to understand your customers and optimize your store.
Continuous Improvement: Keep adding new products, updating your content, and refining your store's design.

Building a successful e-commerce business takes time and effort, but with Shopify as your platform, you have all the tools you need to achieve your goals. Good luck, and happy selling!

Different types of clients I had to face when I was freelance

coco — Tue, 11 Mar 2025 15:13:10 +0000

I worked as freelance developer for 4 years. In those times, I learnt a lot about developing, but I also learnt many soft skills. For example, how can you negotiate the budget? How can you discuss deadlines? How can I communicate goals? How can I understand the features we need for any project? And so on.
However, every client I worked for was different. They had different timing, communications skills, deadlines, versatility and tolerance. Every of each had its own culture.
There are some patterns I have noticed, I list the most typical ones here:

Versatility and tolerance, but a lack of organization

It was one of my first clients. I made landing pages for them. He was a nice guy who I talked to, but he was always in a hurry. Therefore, it was always missing assets, definitions and goals. I started an “admin” once, without knowing the front user design because it wasn’t designed yet. I believe that they hadn’t thought about it yet. We were always adjusting deadlines, and sometimes, something it had planned for March we finally completed in July. Other times you were working on a project, but suddenly you needed to start a another one-more important-and the first one was delayed.
However, they were versatile and tolerant, and they always were in good mood.
They sometimes were late with payments, but it wasn’t bureaucracy, they just forgot. So, I had to give them a call, and they transferred me right away. Other times they made a payment in advance; not even knowing the budget, and they ended up with money in their favor in their account. ‘Keep it for future jobs’, they always said.
Despite their disorganization, I liked working for them. I could understand they were running all the time, and they did always their best to apologize, or to cheer you up.

Extremely detailed

They were super organized. They had their deadlines defined months ago, and even one or two weeks for backup. However, they always became late because they never finish the product. They always had mini changes to the product. ‘What if you change the background color red to a little bit lighter?’ ‘What if you change this text?’ What if you change this photo?’ ‘Put this video instead’. ‘Perhaps, the button should be moved a little bit up’.
I am not saying that kinds of things weren't important, but you need to deploy to figure it out. Once the project is in production, then you can start the next phase, that is the analysis and finally fixing all bugs and making the improvements you want.
If you never go to production, you will never know what your audience thinks about it. If you asked for finishing the project ‘Just one more fix’, they said. But the budget remained the same, and I couldn’t organize the other projects I had, because I never know when I will finish this.
You needed a lot of patience to work with them.

Good money, but it always needs to be done for yesterday.

They had a lot of bureaucracy. The feedback was always late, and they always were adding features at the last moment. Besides, they were not versatile with deadlines. I remember telling them I needed one week to apply all the feedback. However, feedback came two or three days before the deadline, and you needed to run to finish because the deadline didn’t change at all. ‘The president was busy. He could not see it until now. You can charge extra hours, but we need to finish on time’, They used to tell me. Of course, I always did it. I had to run, to work at night, at weekends, or moving other projects to finish it on time.
However, money was very good, and this client made me have an extra to stop for vacations or bonuses. That’s the reason why I liked them. Despite running, I kept them till the end.

Clients have pros and cons. To decide who to work with, you just need to think on the pros and cons and make as a logical decision as you can. There are some clients or projects you would want so much to work with, so your emotions may confuse you. But this is also good. When you like the project you are working on, you stress less. You feel happier, and that is also important. Because you will spend more time working in your life than on anything else. Try to enjoy it!

A failure can teach you more than a ton of courses

coco — Thu, 26 Dec 2024 12:28:34 +0000

Sometimes, you only need to remember what you have done wrong before. What are your big mistakes? What were you sacked for? If you focus on taking some lessons from this failure, you will be better next time. This is called experience.
I remember tons of failures I did, but I want to share with you six of them. For me, the most important:

Complete your task first

When I was a newly developer, I worked as a freelancer. She hired me to make a few adjustments to her web page. I just needed to add a new page and replace one link at the navbar. I downloaded the entire site, for building a local version on my Mac and started developing. It was a WordPress, so it needed to upgrade. She had never asked for it, but I did it anyway. Then I kept working on my to-do list and finished my tasks in just a day. I was very proud and uploaded it. However, it didn’t work at all. I never could find out why. Despite having a backup, I had forgotten to back up the database too, so it was useless. I was desperate. Therefore, I told her what had happened, and I had to develop an entire new site for free. She was really mad, of course, but she loved her new site, and, in the end, she forgave me.
From this experience, I learned the importance of backup, it doesn’t matter if it is a small or a huge fix. You need to do it and test it to make sure it is useful. Besides, I know now that when a project is already started, you are walking on ice. You will never know if the ice is thin or thick. So, you must be careful. A little scratch could become a deep canyon. There is no need to refactor all the time. Sometimes, an upgrade is not necessary either. Sometimes, wisdom is to do only whatever they ask for.

Underestimate project

I was hired to do a small part inside a website. I had to make a tool to design an end of course jacket. The user will be able to choose between any different models, colors, and shapes. Then, they will be typing the name they want printed on the jacket. And finally, they will be sending the design by email. ‘Wonderful, piece of cake’. I thought. I had free time for two weeks, and this job was great to fill up the free space.
I started writing the HTML, it was a frontend job. My idea was to make that the user has as many as jackets' images as combinations possible.
However, it was a disaster. On one hand, the images they gave me were all different sizes. It took me like two days to resize all the images. Because I couldn't only resize them, the jackets had to match each other. And I am not a designer, Photoshop is not my skill.
On the other hand, there were too many combinations. And then, all of them multiplied by 5 or 6 possible colors. It was a spaghetti code. So, I made a matrix with all the combinations, and I put it into a Json file. Each combination = one image.
With the images and the combinations issue, I lost more than 6 days. Remember, I had only 10 days of free time. So, I had only 4 days to finish, and I knew it was impossible. What I thought was easy became harder, and I lost my enthusiasm for the project. The results were more than double of my time, working after hours. Because I underestimated it. I should have taken my time to analyze the project to have a better understanding of it.

Too big project

Two years ago, we started estimating a new project. We were very excited, because the team hadn’t been doing anything important that year, only intern projects, and it was a huge opportunity and a good challenge. React JS, Node JS, MySQL, and Redis for cache. I remember saying ‘6 months’ without thinking too much. Then, we sit to negotiate. We finished estimating in less than 4 months, and the project took more than 7 months to be completed.
The client knew exactly what he wanted: A revamp, better performance, and new technology. However, they didn’t know any details about anything. Any question we did, it was very difficult to find a person who knew the answer. Reading the old code, we discovered many functionalities the client didn’t know existed. Besides, the client was very happy with the new design, but developing it became hard.
The team had never worked on such a lengthy project before. We were used to short ones, 2 or 3 weeks. The estimation was so bad, but the methodology we chose was worse. We designed all first, and then we started developing. The project was too big for doing that. We should have designed a little bit, started developing this part, checked with the client and went on. Agile methodology would have been the correct choice.

Incorrect Focus

In my other life, I was a film director. I won a prize to make a short film for Buenos Aires city. I was so proud of my script, it had zombies, romance, and a message. The prize was money to film it. So, I reach a group of friends to help me do it. I did a casting, and we were ready to go. I used to live with a make-up artist, and she made up as zombies one time to go to a Halloween party. We were the party sensation. However, I told her to be the art director. She said yes, but I should have asked her to do the short film zombies. I hired a special effects company instead. It was the most expensive item in the budget. Therefore, I had to shortcut in all the other things.
The zombie was great, but it wasn’t the most important part of the film. My friend could have done the same, and with less money. I was wrong. Because I had to cut the days of filming and run, reducing the quality of the shots. Perhaps the short film would have been better if I was put the focus where it mattered.

Too many people

When I became the Director software development at Portinos, there were only three of us. I wanted to make the area grow. The first year we grew 100%. It was great, I was very happy then. However, there were only six of us. Therefore, the second year we continued growing. Suddenly, when there were over than ten of us, I noticed I made an issue. Too many people in the room, I couldn't handle them. The result was that many people quit, and I had to restructure the area. If you are going to grow, you will need more middle managers, so you won’t lose the balance.
Nowadays, we are much better. There are sixteen of us, and there are two teams with their own leader. This year we are going to create a new team also. We keep walking!

Quit without thinking

When I was studying to be a film director, I started working at Ogilvy. The multinational and famous advertising agency. I worked there for one year. My job was to make copies of the advertisements and send them to TV channels. In those days the distribution was made on video tapes (Beta SP and Umatic), so I took the master and copied many times as needed. It was a repetitive task. However, sometimes it was fun, because the creatives (the people who think advertisement) needed to find references to tell their ideas better. I was in charge of the Ogilvy and other agencies’ advertisement catalogues. So, the creatives used to come down to talk with me, and we spent many hours making a reel to show to the client.
In those days, I wanted to be a film director, and I finished quitting the agency because the advertisement was too frivolous for me. I wanted to make films, indie films as possible. I asked for a salary rise, my boss denied it, and I quit.
I know now that I should have been kept working there. Advertisement is a job as any other, and it is also an art. In only 15 seconds you need to tell story, that is quite difficult, and a huge challenge. But I was only 20 years old. I didn’t really need the salary rise. It was just an excuse. I didn’t know how to negotiate it as well.
This was the biggest regret in my life. However, it was a fork in the road, and I chose my path. Next time I should stop and think a bit more before starting to walk.

These are the most important failures, or at least, the ones which I remember most. Why do things right when you can do them wrong and twice? :D

Advergames

coco — Tue, 17 Dec 2024 14:51:52 +0000

When I was a child, a video game was only for kids. Nowadays, there are more adults playing than children. There are even people who are professional players. This means that they earn a salary for playing video games like a football player or NBA basketball player. League of legends’ final had more than 7 million people watching the match in November 2024. Esports are now the icing on the cake.

More than 3 billion active players around the world are enough to pay attention. There is a huge audience. Besides, more than 50% of them are adults; they have their own money and could buy whatever product they want without asking permission from their parents. For this reason, it is very important for brands to consider that video games are a new channel for advertisement. Could be for selling their products or just to talk to their audience. Esports are the new sports as streaming is the new television.
Therefore, the advertising industry has already noticed this; and many experiences were made. Depending on the brand’s target is the video game to create for. Kids are playing on Roblox, casual adults are on mobile, advanced gamers are on PC or consoles, etc. It will also be important the genre of it: Horror, strategy, sandbox, shooters, sports, puzzles, etc. Every genre has its own audience. Besides, it is not the same as creating a horror game or a shooter than a sandbox or a puzzle. The implicit message the brand is giving is not the same. For this reason, it is very important to consider what do we want to communicate, and the campaign’s final objectives.

Advergames are used for almost everything now. Nikeland is selling t-shirts (NFT), but other companies are there only to communicate their values or remind the audience of their vision. Surveys were also made or even focus groups. Imagination is the limit.
Besides this, nowadays, is easy to do it. In the 90’s, if you wanted to create a video game you would need to start from the scratch. We are full of information and different technologies to use now. However, the most important is that you can use platforms or technologies already made up, and the only thing you need is to think of a good idea. For example: Roblox is a 3D render motor, but it is also a community, a marketplace, and a framework. You only need an idea and Roblox makes easy for you to develop it. Roblox also has a marketplace already made, a community to share your game with, scripts, and 3D models for free. It so easy to develop and publish a game there, that young kids are doing it. Minecraft is another example; Fortnite is doing the same thing as Roblox did.

To communicate is the challenge.

There are so many video games now that it is difficult to stand out. However, if you know what you want to tell your audience and which is your objective, will be ok.
It is important that your game and your marketing campaign are aligned because the brand and the game must be the same. Considering your video game as a branch. When you open a new branch, you always need to communicate, to advertise and to be patient. If you did the market research well, you wouldn’t need to worry. The sooner or later your audience will go to the new branch. This is the same! Let the video game to grow. But never stop communicating it.

Therefore, Advergames are a new opportunity to communicate your brand, talk to your audience, and even sell. It is marketing, PR and commercial in the same channel. Nowadays, is easier to develop and publish, but it is difficult to stand out. However, video or graphic advertising is even more saturated. I think this is a new opportunity to take advantage of. Let’s play!

Jammin’

coco — Tue, 17 Dec 2024 14:48:15 +0000

All my bags are packed
I'm ready to go
I'm standin’ here outside your door
I hate to wake you up to say goodbye
But the dawn is breakin’
It's early morn
The taxi's waitin’
He's blowin’ his horn
Already I'm so lonesome
I could die

Penso che un sogno così non ritorni mai più
Mi dipingevo le mani e la faccia di blu
Poi d'improvviso venivo dal vento rapito
E incominciavo a volare nel cielo infinito

Au-dessus des vieux volcans
Glissent des ailes sous le tapis du vent
Voyage, voyage
Éternellement
De nuages en marécages
De vent d'Espagne en pluie d'Équateur
Voyage, voyage
Vol dans les hauteurs
Au-dessus des capitales
Des idées fatales
Regarde l'océan

We're jammin’
I wanna jam it wid you
We're jammin’, jammin’,
And I hope you like jammin’, too

The song Jammin by Bob Marley, has the simplest lyrics, but it is also the most beautiful song, and the one that lets you feel the vacations’ spirit. Reggae is the kind of music you only need to relax. You listen and feel free. In Jamaican slang, the word jamming also refers to dancing very close together during a celebration to the rhythm of music. This is all what a holiday is about. Relaxing and letting you go!

“I wanna jam it wid you, and I hope you like jammin’, too.”

Vacations are very important for your health, your mind, and your spirit. There are many studies that talk about it. Benefits for your mental and physical health, personal relationship, productivity, and motivation. Vacations reduce the stress and improve your mood. That’s why it enhances creativity and increases focus.

You don’t need to travel. I took many vacations on my sofa that were amazing. I loved walking by the city without going anywhere, coming back home late, or just lying listening to music while looking at the ceiling. I didn’t have much money to travel, but I used to love my vacations, and I could rest the same.
However, it is important to relax and jam. If you thought about what you must do when you come back to work, it wouldn’t be useful at all. If you talked with your team in those status meetings, it wouldn’t work. And if you chose to do something similar to your job’s activities (it doesn’t matter if they are your own stuff), it wouldn’t be restful enough.
Whatever activity you choose, it must be as different from your daily routine as possible. Besides, you should enjoy it. :D
Another thing I love is to disconnect. I always leave my iPad and Mac at home. Nowadays, it is impossible not to take the phone, I even have my wallet on it, but I uninstall work apps such as Slack, Jira, and Outlook.

We have many holidays in Argentina. Christmas, Easter week, Workers Day, Independence Day (we have two days for this), and many others. I like these mini vacations we had, because we can relax a little bit during the year. Science says that you need at least two weeks for relaxing your mind and body. Thus, vacations are very important. However, I use these mini vacations for relaxing. Our parents don’t live in the city, one family live 90 minutes’ drive and the other family 4 hour drive. Therefore, these few days are great for visiting them.
Walking around the city could be another good option. I love to be a tourist in my city. Thus, these mini vacations are not even one week, but they are very useful for burnout prevention.

I think I have never interrupted a vacation to anybody. It is important as a team to work on developing that policy. Not to send any messages after work hours, and not to call an employee who´s on a vacation.

Reset is the first step in almost any troubleshooting. Let our body reset -at least- once a year it is important. CTRL + ALT + DEL!

https://youtu.be/oFRbZJXjWIA?si=aGiT5G4WPLH37wpm

Round of applause

coco — Tue, 26 Nov 2024 15:18:04 +0000

December, end of the year. It is the time of reflection and setting new goals or objectives for next year. Wait! How many posts of goals-2025 have you read? I don’t want to add one more of that. I want to bring a different approach instead. What if we start celebrating for what we have done during 2024. Could be so little or very much, it doesn’t matter. There is always something to celebrate about.

Don’t let me be misunderstood. It is important to set new goals and continue moving forward, following your vision. I am fan of planning. I will finish doing 3 or 4 scenarios for next year. However, it is also important to stop and celebrate too.

The routine keeps us moving forward, and we sometimes forget to celebrate our successes. This is as much important as planning. It is not necessary to throw a full party. If it is only a speech, then is also useful. The important matter is taking the time to stop and celebrate. We have made a huge step doing only that.

The other day, I was talking with my boss in my 1:1, and he mentioned that he was happy because we finally got that project, we have been following for 6 months, and it was a huge success. However, he couldn’t say a word because when he was notified, he was attending an important meeting. Therefore, he asked me to make a call with the team just to celebrate.
The idea was telling the team about the importance of the project and that he was very proud and happy with them.
The team was delighted. I think that kind of actions are so great because people feel heard and supported by the company.

Despite this meeting, I couldn’t help thinking about all the successes we had during the year, and we didn’t celebrate. As I wrote before, the routine keeps us moving forward. We get a new project, complete it, start a another, and so on. Hence, next year my first goal will be celebrating every step we will be taking.

What if you take the time to think about how many steps have you been taking during the year? What are your achievements? How many goals have you reached? And the most important, what have you learned?

Remember that you learn from your victories, but you learn more from your failures. Therefore, you should have learnt something. Only for that, round of applause!

If you want to be successful, you should stop for lunch!

coco — Wed, 23 Oct 2024 15:15:05 +0000

I mean it!

Everybody knows nap’s benefits. It relaxes your brain, reduces fatigue, increases alertness, and makes your day. Therefore, your performance is improved by one hundred percent.
However, nowadays, the truth is that we can’t take any naps at all. Not in our lives!
I live in Buenos Aires, a big city where everybody goes to work every day in the morning and comes back home in the evening. Nobody has a place to sleep in their offices. Although, I do home office, I can’t take a nap either. That’s why I have only one hour for lunchtime, and that is not enough. It is very difficult to cook, eat, and have time for sleeping as well. If I try to do everything, I will finish doing nothing at all.
Nevertheless, we can stop for lunch. We must do it. Because lunchtime is the new naptime.

In the Agency where I work, everybody does home office. Our time for lunch is only 1 hour, between 12 pm to 3 pm. However, I see many people are not taking advantage of this hour. Perhaps, they eat something while working; they eat sitting at the computer while surfing the web, or -the worst-, they eat something precooked in 5 minutes. I’m pretty sure everybody is having a date with their cell phones instead of having lunch.

As a result, having lunch like this isn’t worthwhile.

What is the alternative?

Lunch is one or of (or even) the most important meals of the day. In Argentina we are used to have bad breakfasts: we don’t eat eggs, vegetables, or meat as in other countries. For this reason, lunch is an opportunity to provide our bodies with the nutritious they require.

Moreover, having lunch is to have a break. You need to stop and relax. Do nothing for a moment or do something completely different from work. Only in those cases, your brain will be free, rested, and ready for the next hours. Otherwise, if you need to cook lunch, it will be ok as well. Because the brain will need to be focusing on cooking, and this is a completely different activity unless you were a chef. It may sound silly, but this rest period for the brain is critical. Reset!

If you can’t take a nap, you must stop for lunch. Therefore, you will feel de-stressed day by day, your brain will improve focus, and your health will be better. This is the first step to be successful.

Now, junior, behave yourself!

coco — Wed, 11 Sep 2024 15:49:02 +0000

September 12th. Today is the 256th day of the year. This means it is Programmer’s Day! 8 years ago, was the first day I did celebrate it, because I had been hired to do my first big project, on which I worked for more than two years. I remember feeling excited and proud. Especially proud! However, I was also reckless. I was hired to do something, I didn’t know how to do, but I knew I would be successful in the end. Of course, when they asked me if I knew, I immediately answered ‘yes, of course’.
That project got me into a path of many headaches, a little bit of burnout, and especially infinite coding hours. That was a huge learning path as well, and an inversion too. I think that I was being almost a senior dev, despite being junior. Therefore, as today is programmer’s days, I would like to give some advice for junior’s developers.

Being a software developer.

First, I would like to say that I live in Argentina, Latam. This is important, because I don’t know how being a developer in another country or region and these bullet points, I am writing below, may be different or not in other locations.

Second, this is what I think or believe because it is my life. Take what you want, and then readapt them for your own context. Sometimes what is good for someone is not for another person. There is a lot of advice over there and sometimes it is similar, sometimes is not. Again, I believe this today, but this is not math, you could do great as well if you won’t take any of them.

Third and finally, don’t re-invent the wheel. Try to take as much advice as possible. The more, the merrier. I know this is contradictory with the second one I wrote above, but that’s how life is.

What does a software developer do?

A software developer is a person who solves problems. If you want to be an artist, be an artist, not a developer.

What do I mean by this?

You need to focus on which problem you are solving. It is always one. It can be hidden or explicit. Sometimes, nobody is thinking about it, and the brief is about something else too. This is because it is difficult, as a person, to explain us, and sometimes also to know what we want. In our cases, the brief is usually made by account person who is listening to other people who don’t know exactly what they want, and the account are doing their best to make a brief. However, take my word, there is always a problem.
The issue sometimes comes from another team member. For example: You need to adapt that beautiful design to many screens, browsers, and devices in the world.

If you are solving a problem, it must be solved as soon as possible. Therefore, you must always keep as simple as you can. Because simple is faster and cheaper. That’s the main reason you need to focus on the main issue you are fixing now. If you are doing a landing page, perhaps, you don’t need to use React, NextJS or WordPress. Unless using one of these technologies makes you work -at least- 2x fast.

How can we be such simple?

DRY is a common principle in software development that reads, ‘Don’t repeat yourself’.
For example: if you are writing a program which calculates the distance between an element and another, a good practice will be to write a function to re-use it any time you need to calculate this.

However, there is another famous software principle: KISS. ‘Keep it simple stupid’.
If you need to calculate this only once, there is no need in making a function. Because you are making the code more complicated.

These two software principles show that being simple is the hardest thing ever. Anyhow, the only way to be simple is learning and gaining experience. Therefore, a good software developer must always be getting new techniques, new technologies, and especially new skills.

A new technique could be useful to make the same job faster. But it is not enough just to know it. Practice, and practice more, will help you to get better in your daily work.
A new technology could be useful to solve a problem in a different way. There are sometimes faster, sometimes slower, but there are always two or more solutions. Don’t forget: Technologies are just tools. Thus, if you have many tools in your toolbox, you will work faster, easier, and better.
Developers must develop new technical skills. If they are backends, they should learn frontend and the other way around. A developer must know about hosting, servers, domains, and networking. Not for changing their jobs, but it will be helpful just to know what happens when their job is finished. If they are making the front, it is useful to know where the data comes from. If they are making the backend, it will be useful to see how the user will interact with all the processed data. Besides, you will need to know how the data flows over the internet as well.
However, everybody should be working on what they want. I am not trying to push everybody to do everything. I wrote before, you should only know. Because, sometimes, knowledge is enough. A developer mustn’t be an octopus but must be open-minded. “Curious” is the magic word. The more they read, watch, or listen to; the better prepared they will be.

We could learn all this by ourselves, though the best way to do it is by sharing experience with other people: our partners, but also designers, stakeholders, everybody…

Networking!

Soft skills are important too, don’t forget about it. Many people were born with soft skills included, but others, like myself, have to practice them a lot.

Good developers know how to express new ideas, explain difficult tasks, or tell a partner what we need to do.
We should learn to listen behind the words to detect what the problem is we need to solve.

It is useful to chat with everybody: Designers, stakeholders, account manager, QAs; ask many questions as you can. Everybody has a different point of view, and this is useful for us to see the whole picture.

And the most important: We must always transfer knowledge to new developers. Because someone helped us before, and we have to pay it forward.

Being a team player. This is the way!