Ivo De Rosa

How to Build a Portfolio

Tue, 01 Sep 2026 08:00:00 +0200

Many designers, friends and students alike, have asked me since forever how to build a portfolio (and where), and consequently how I built mine. Which tools did I use. How long did it take me. Should I use Wix? WordPress? Behance? And I always say the same:

It’s complicated.

Because I have strong opinions. And I hate to repeat myself. So here we are.

Perhaps you’re one of these people, or maybe you’re a complete stranger (how the hell did you get here?), but in any case, here’s what you’ll find here:

Arguments to build your own stuff. Free technologies to do so. Someone to talk about it. Motivation. Comradery.

No tutorials nor tricks to sell you anything. No BS copywriting.

This is the first post of this brand new blog, so it only makes sense to explain how the thing was built and what I learned on the process. And what can you get out of it, too.

The short answer

Jekyll + GitHub Pages + Obsidian . 12 bucks per year for a domain registered on NameCheap¹.

Just tell that to your stupid AI agents and they’ll build it for you. But that’s not what we’re here for.

In short, Jekyll is a site generator. A landing page, as you may already know, is generally built with some flavors of HTML, CSS and JavaScript.

Jekyll uses all that to turn your Markdown files into a site, and it’s surprisingly easy to set up, even for a humble designer like yours truly.

Those Markdown files can be created in basically any tool, but I use Obsidian as my text editor and PKM (or however the hell we call it these days)

I can use Obsidian from my Mac, iPhone and iPad. I’m actually writing this on my phone, from my couch. For free. Isn’t that cool?

But why these tools and not others!?

Why Obsidian? Because it’s free and I own my files. And this feature is absolutely wild:

Why Jekyll? Cause it works. And I like it.

Why am I not using a newer, flashier framework/tool?

Who cares. Really.

This is like the Sketch v Figma debate from a few years ago. Just use what’s best for you/your team/your context.

You’re not going to be “left behind”, specially if you’re a designer.

I’ve seen two dozens of tools that look like Jekyll, work like Jekyll and are better than Jekyll for x or y reason: Astro, Gatsby, Next.js, Hugo, Eleventy… new stuff comes out all the time.

Pick the one that looks less daunting to you and crack on. There’s always time to migrate or to learn new stuff down the road if needed. Our goal now is to build something to put our stuff out there.

GitHub Pages was built for Jekyll, so that’s one less problem.

Owning instead of renting

But why not just use Wix or WordPress or Behance, then?

Well, I like to build things my way. To make them look exactly how I want them to look. To understand how they work. To express myself.

This is my little corner of the Internet, after all.

And because I hate paying a monthly fee for stuff I can build myself while having fun, too.

Don’t know how to code? Neither do I. I just read the docs. Buy a template. Break things. You can ask AI for help, sure, but we’ll miss the point if we ask it to build it for us.

There was no AI when I first built this site a few years ago under a different, long forgotten name. It’s doable. It only takes patience and some spare time.

I want to be fair here: Wix and Squarespace and the rest are perfectly valid choices, especially if you don’t have a technical background. They get you from zero to a live site in a couple hours.

But if you do work in tech (or are planning to do so) without a technical background, what a wonderful excuse to start building one.

I understand that convenience is quite valuable in the era where in order to even be considered hireable you need to be (or at least appear to be) productive all the time and constantly putting things out there.

And I say this as a father and a worker. I know. Convenience is valuable indeed, but it can also harm you in the longer term, because it’s a doubled-edged sword.

One of its side effects is called deskilling. So be careful.

Building and maintaining your own site, even slowly, is how you hold on to it. Today I might be able to build this whole thing from scratch without a template, because I’ve spent enough time messing with it.

The time you invest in perfectioning it, fixing its bugs, creating with(in) it is not wasted, it’s literally the opposite (I said investment on the first place didn’t I)

The backbone

This site is built on Clancy, a Jekyll theme built by a guy called Artem Sheludko. It comes packed with features for only ~80 bucks.

Setting up the backbone of a project (the grid, the type system, the responsive behavior…) sucks.

Of course I want a responsive typography system, but I’d rather “just” define the values I want to use than doing that and also write classes for every style on every breakpoint.

Also opening that theme was like opening a good Figma file where you first learn how to nest components, or that good old (now obsolete) :slot trick for the very first time.

I love reverse engineering stuff, man. I really do.

No AI at all then?

I mean, I burned one billion tokens this year alone. I routinely send 10 minute voice notes to my Hermes agent to structure my thinking. I ask it to audit my content, to do research for me… to do automated QA when I develop something new for my coffee app, too.

So yes, I use it now to tinker with this site.

I’m not saying don’t use AI, I’m saying be careful. Because it’s convenient. It literally makes you dumber. And the companies spearheading its expansion are doing horrendous things.

I believe in a different model, but that’s for another post.

My point for now is: be smarter. Tinker. Learn. I trust in you.

I’m dead serious.

The non technical part

This is called how to build a portfolio, but so far we’ve only discussed tools, tech and philosophy; nothing about strategy or content design yet.

Let’s add a bit of that too.

Pick your audience

Think about who actually reads your portfolio. For most of us, it isn’t another designer admiring our craft (are you? I’m blushing) — it’s a hiring manager, skimming twenty portfolios in a morning, looking for a reason to call you back.

So choose what you add for them. If the site is meant to convert (to get you a conversation) lead with the thing that earns that conversation, not the thing that proves you’re clever.

And decide deliberately what belongs on it and what doesn’t. Do you want a section for experiments and half-finished things? I do: I call mine Playground, and it’s there for other tinkerers too.

Just know why it’s there, and who it’s for.

Oh, and if your portfolio is not for hiring managers (or not only for them), this advice is equally valid, because what I’m saying is: design for the audience you want to reach, whatever that is.

Sounds obvious, but we often forget it, so there’s no harm in repeating it a few more times.

This blog exists for me to sharpen my thinking (and maybe help you out in the process too), and it’s part of my Portfolio cause for me it’s all one and the same.

My writing skills are, in a way, my design skills.

Show don’t tell

There might be a shift in how designers are presenting their work right now, and in what the market demands.

For years the formula was: narrate the process, discovery, personas, wireframes, iteration… It’s what schools and bootcamps have taught for a decade, and it’s why so many portfolios read as completely interchangeable.

And a big chunk of that was fake. I’ve hired designers. I’ve been a teacher. I know y’all lie. It’s ok.

Well, it was.

What you hear now is: what did you build, how, why, and what ideas do you still have?

That’s in part why I wrote this. Launching some crappy app with Lovable and get some real world usage data might look impressive for a while, but the real thing that will stay with you forever is starting to wrangle with the terminal right now.

I’ll never forget the day my friend Carlos saw me frustrated trying to use XAMPP and said: hey, why don’t you use npm run dev?

That said, I understand the pressure to make your projects look successful, but this is like dancing: if you focus on the joy of doing it, people will see your passion and tenacity, not only the fact that you’re just pretending to know how to dance salsa (meaning: your project sucks)

Mistakes will just be a natural part of it, because their lacking in case studies has always been a bit suspicious, if you ask me.

One last thing

I got really good advice from my friend Helena a few years ago, and I want to share it with you too.

Everyone knows you have to get past the software that filters résumés — the ATS, the keyword scanner, the gate before a human ever looks.

But optimizing only for that, turning your CV or your site into something surgical and professional and entirely free of personality, is missing the point.

She told me to put a bit of myself into it (not literally, for God’s sake): the fun facts, the family, the random bits. The little things that have nothing to do with “professional experience” and everything to do with whether someone actually wants to work with you or not.

Designers forget this, because we often treat content like something clinical. You want to sound and look like Apple, but turns out you’re a dude (yes, I’m talking to myself)

Your portfolio doesn’t have to be a sterile, official document. It’s also a place where you get to express yourself freely!

And I too hope I get mine to look more like me, step by step. In Spanish we say “consejos vendo, pa’ mi no tengo”

Now, back to designing things. More posts soon.

That’s what I pay the Internet cartel(s) for my domain ↩

Letter to my 20-year-old self: On Work, Care, and the Myth of Automation

Mon, 11 May 2026 00:00:00 +0200

Dear 20-year-old Ivo,

I know where you are right now. You’re probably sitting there, convinced that we are on the precipice of a Great Liberation. You’ve believed it since you were six or seven—the idea that machines will eventually do everything for us. That “work” is a bug in the human experience, a relic of a primitive era that automation will simply erase. You’re not just hoping for a shorter work week; you’re waiting for the end of labor itself.

I hate to be the one to tell you this, but you were wrong. Not about the technology—the machines are here, and they are faster and “smarter” than you imagined—but you were wrong about what work actually is.

For a long time, you confused labor with work. You hated the wage slavery, the grind, the selling of hours for survival. And believe me, I still do. But in your eagerness to automate the boredom, you forgot to account for the value of presence.

Today, I had two visits. A midwife came to the house in the morning, and I went to the dentist in the afternoon.

Think about those two things.

A midwife bringing a human into the world. A dentist working millimeters away from your nerve endings. In both those rooms, the last thing I wanted was a “technically correct” machine. I wanted a human. Why? Because a machine optimizes for a result; a human optimizes for a person.

There is a specific kind of judgment—a criteria of “good”—that only a human can apply. It’s the ability to tell if a decision is right for you based on your preferences, your fear, your breath, and the subtle energy in the room. Not just whether it fits a statistical model of “success.” That expertise, that embodied knowledge, is where the real value lies. It’s not about efficiency; it’s about care.

And here is the real kicker, the part that would make you furious: the cost of the “liberation” you’re dreaming of.

Right now, there are supercomputers like Colossus 1—monsters built by the likes of Musk and used by companies like Anthropic—that are literally poisoning the water and air in places like Memphis just to keep the servers cool. We are destroying the physical world, the actual earth we walk on, to build a digital proxy that is, quite frankly, mediocre.

We are trading the health of our planet for a technology that can write a decent email but can’t hold a hand or fix a tooth without a human overseeing every single second of it. The future they are aiming for—this sterilized, automated utopia—is fucking bullshit.

So, if I could give you one piece of advice: stop waiting for the machines to save you from work. Instead, start looking for the work that is actually worth doing. The kind of work that requires a body, a soul, and a level of expertise that cannot be reduced to a prompt.

Don’t fight against work. Fight for the dignity of human mastery. Because in a world full of mediocre AI, the only thing that will actually matter is the person who knows how to be human in the room.

With love (and a bit of a toothache),

Older Ivo

[!NOTE] Editorial brief Idea: letter from Older Ivo debunking the “automation will free us” belief he held since age 6–7. Core argument: confused labor with work; real work = care, presence, human judgment a machine can’t reproduce. Anchors: midwife (morning) + dentist (afternoon); Colossus 1 poisoning Memphis. Origin: dictated as voice note 11 May 2026 (found again in chat history). Audience: AI-native HMs/devs/PMs — interrogates the “automate everything” orthodoxy. 🔵 AI·work·skill (thesis). Status: essentially complete — strong publish candidate. Bridge to parto-en-espana-diseno-servicios (same midwife scene).

The Craft of Taste: Serving the Material

Sun, 10 May 2026 21:00:00 +0200

In an era of mass production and rapid consumption, the concepts of taste and craftsmanship often feel like an indulgent luxury rather than a foundational principle. Yet, true artistry, whether in filmmaking, writing, or any creative endeavor, fundamentally relies on these twin pillars. Alfonso Cuarón’s approach to directing Harry Potter and the Prisoner of Azkaban offers a masterclass in this philosophy, a lesson that resonates deeply with any process of drafting and refining ideas.

When Cuarón took the helm of the third Harry Potter film, he was presented with a massive, beloved franchise. His mentor, Guillermo del Toro, offered invaluable advice: “Don’t try to do your thing. Just try to serve the material… And if you serve the material, you may do your most personal film.” This wasn’t about stifling creativity, but about a profound respect for the source — an act of humility that paradoxically unlocked a deeper, more authentic expression.

Cuarón’s success lay in his “thematically driven adaptation.” Instead of merely translating plot points, he identified the core emotional truth of the book: Harry’s journey of identity and of confronting inner fears. This keen editorial taste allowed him to make purposeful choices, grounding the fantastical elements in human emotions. The goal was “a movie that was character-oriented with a special effect, rather than to be a special effect film with a bunch of characters around.”

Every decision — from the moodier cinematography to the subtle shifts in lighting, and even allowing the young actors to wear casual clothes to reflect individual growth — was in service of this thematic vision. His collaborative spirit, involving actors in character development through essays, further enriched the final product. It is a testament to craftsmanship when every small detail aligns with the overarching narrative and emotional arc.

Just as Cuarón built a pivotal framework for the entire Harry Potter saga by foreseeing its overarching themes, any creative process demands the same dedication: keeping a detailed record of raw ideas and drafts, sifting through references, and structuring an argument so that every sentence serves the core message. This slow, deliberate refinement — separating what truly matters from what merely fills space — is the craft itself.

Ultimately, the act of creation, whether crafting a blockbuster film or a concise essay, demands both a discerning taste to identify what truly matters and the meticulous craftsmanship to bring it to life meaningfully. By serving the material with integrity and purpose, we not only honour the source but also imbue our work with a personal authenticity that resonates with our audience.

[!NOTE] Editorial brief Idea: taste + craftsmanship via Cuarón serving the material. De-meta’d: removed the internal file references (draft_references.csv, Channel Plan.md, etc.). Audience: 🟣 life·place / craft. Good tier-2 piece. Status: ready after de-meta.

Childbirth in Spain: a personal account and a service-design analysis

Sun, 10 May 2026 21:00:00 +0200

Introduction

Our daughter hasn’t been born yet, but the months of preparation have plunged us into a world of information, emotions and, above all, the realisation of a childbirth system that, despite its technical advances, shows serious flaws in its human-centred design.

This post comes from the need to document our journey, to turn unease into action, and to offer constructive criticism from two perspectives: the one lived by those of us who are about to become parents, and the one that comes from my background and experience in service design and health technology (Tucuvi).

1. The numbers that hurt: beyond the caesarean rate

We are told that in Spain the caesarean rate sits at around 26.4% (Ministry of Health data, 2024)¹, far above the 10–15% range recommended by the World Health Organization (WHO)². But that number hides individual stories.

Behind every percentage there is a woman who perhaps did not get the chance to give birth the way she wanted, who underwent an intervention without her full consent, or who felt that her body was taken over by protocol before her own physiological process.

An investigation by elDiario.es revealed that 26 public and private hospitals exceed a 45% caesarean rate³ — that is, nearly half of the births in those centres end in the operating theatre. Some of the examples cited include:

Hospital Universitario de Guadalajara (Castilla-La Mancha)
Hospital Universitario Virgen de la Arrixaca (Murcia)
Hospital Universitario de Burgos (Castilla y León)
Hospital Universitario de Salamanca (Castilla y León)

2. Obstetric violence and the weight of words

While preparing for our birth, we came across the term obstetric violence. Its use is controversial because many healthcare professionals perceive it as a personal accusation rather than a sign of systemic failures.

Yet the evidence is overwhelming: practices such as the Kristeller manoeuvre (fundal pressure during the pushing stage) are still used even though they are advised against by the WHO, the Spanish Society of Gynaecology and Obstetrics (SEGO) and the Ministry of Health itself⁴. This manoeuvre can cause uterine rupture, severe haemorrhage and fetal damage.

The debate over terminology reveals an ideological bias in the leadership of some services. In Murcia, for example, it has been reported that the conscientious objection of the head of the Obstetrics and Gynaecology department blocks access to legal abortion⁵. Although we are still verifying this specific case, similar patterns are seen in other regions where the ideological charge of clinical leadership determines which protocols are applied and which rights are respected.

3. Service design: what I learned at Tucuvi and in Health Design Thinking

My work at Tucuvi taught me that a well-designed medical protocol can save lives, but that a rigid, poorly adapted protocol becomes a machine for producing unwanted outcomes. In childbirth, we see how:

Excessive standardisation removes the ability to adapt to individual needs.
The lack of feedback loops prevents women’s experiences from improving the protocols.
Hierarchical power (department heads) freezes innovation; some hospitals’ birth plans haven’t been updated in decades (it is rumoured that at Los Arcos, Murcia, the plan has been in force for more than 20 years)⁶.

Contrast this with the United Kingdom’s model, where the clinical best-practice guidelines (NICE) are reviewed every 3–5 years, forcing hospitals to renew their protocols⁷.

4. “La voz de las mujeres” and the activism that guides us

The documentary “La voz de las mujeres” by El Parto es Nuestro sums up twenty years of activism against obstetric violence⁸. Its milestones include:

Bringing unnecessary and harmful manoeuvres to light.
Promoting the birth plan as a tool for negotiation and informed consent.
Campaigns such as “Stop Kristeller” and “Haloperidol en el parto jamás más”.

5. Our choice: Icosaedra and the home birth

Given this landscape, we decided to hire Icosaedra Maternidad to accompany a home birth⁹. This decision is not a rejection of medicine, but an assertion that we want a service where:

The woman is at the centre of decision-making.
The protocol adapts to the process, not the other way around.
The physiological time of birth is respected.

Conclusion: design for birth, not for intervention

Childbirth in Spain needs a design revolution. It is not about eliminating medical technology, but about putting it at the service of people. To that end we propose:

Mandatory transparency of intervention rates by hospital (as elDiario.es already does).
Periodic updating of protocols (every 5 years at most).
Mandatory training in human rights and informed consent for all healthcare staff.
Bringing midwives into the decision-making of obstetrics services.
Validation of birth plans as binding documents between the user and the service.

Our son or daughter deserves to be born into a system that respects their process from the very first heartbeat. And we, as future parents and as designers, have the responsibility to demand that change.

This is a draft. We will publish the final version after the birth, incorporating our direct experience and the lessons learned.

[!NOTE] Editorial brief Idea: el parto en España desde la experiencia personal + el diseño de servicios (Tucuvi). Cesáreas 26,4% vs OMS; violencia obstétrica; Kristeller. Audience: designers (service-design hook) + anyone touched by healthcare. 🟣 life·place but bridges to career. Status: unfinished — “todavía no ha nacido nuestra hija”. Finish the closing + references before publishing.

Ministerio de Sanidad, “Informe Anual del Sistema Nacional de Salud 2024”, https://www.mscbs.gob.es/estadEstudios/estadisticas/ informesAnuales/SNS/2024.htm (accessed May 2026). ↩
Organización Mundial de la Salud, “WHO calls for global expansion of midwifery models of care”, 18 June 2025, https://www.who.int/es/news-room/detail/18-06-2025-who-calls-for-global-expansion-of-midwifery-models-of-care ↩

Raúl Sánchez and Ana Requena Aguilar, “BUSCADOR

Consulta la tasa de partos por cesárea de cada hospital de España”, elDiario.es, 12 October 2025, https://www.eldiario.es/sociedad/buscador-consulta-tasa-partos-cesarea-hospital-espana_1_9622470.html

↩

El Parto es Nuestro, “Dossier STOP Kristeller: evidencia científica”, https://www.elpartoesnuestro.es/informacion/dossier-stop-kristeller-evidencia-cientifica ↩
Note: verify with local sources (e.g. Murciasalud, reproductive-rights platforms). This claim requires confirmation through archive research and official statements. ↩
Source needed: direct enquiry to Hospital Universitario Los Arcos or a review of its public protocol document. ↩
National Institute for Health and Care Excellence (NICE), “Caesarean section”, last updated 2023, https://www.nice.org.uk/guidance/ng192 ↩
El Parto es Nuestro, “La voz de las mujeres”, https://lavozdelasmujeres.elpartoesnuestro.es/ ↩
Icosaedra Maternidad, “Acompañamiento Parto en casa”, internal document provided by the company (April 2026). ↩

Your portfolio doesn't have to weigh 150 MB

Wed, 23 Sep 2026 01:33:36 +0200

My portfolio repo had 146 MB of images. Two of them were 22% of the total. One was a PNG 11,811 pixels wide.

I only found out because I finally measured instead of guessing. This is what I learned, with the actual numbers — and the rules I now apply to every image I publish.

First: the repo is not the page

The instinct is to panic about the total. Don’t. What a visitor downloads is what a single page references:

Home: 1.27 MB across 3 images
Work index: 2.67 MB across 7 images
A case study page: 0.02 MB in one image

Nobody was ever downloading 146 MB. But the repo size was still a symptom of the real problem.

The real problem: originals that were never exported

The images weren’t “heavy”. They were raw exports:

arbor_cnp_2.png — 23.4 MB, 11,811 × 8,268 px. Nothing on the site is ever displayed wider than ~1,500 px.
arbor_gif.gif — 13.3 MB (600 × 450, 13 seconds)
arbor_gif-2.gif — 18.3 MB (600 × 600, 5 seconds)

Two GIFs, 31.6 MB. That’s the whole problem in two files.

Rule 1 — pick the width from the layout, not from the file

The right width is display width × 2 (retina), and it changes with the role of the image. In my site:

Role	Displayed at	Export at
Grid card (3-column, 1140px container)	~370 px	800 px
Image inside a case (740–760px column)	≤760 px	1500 px
Full-bleed / hero	viewport	2000 px

“Export everything at 1800px” is what I did first, and it’s wrong: it wastes weight on cards and falls short if something opens full-screen. If an image appears at several sizes, ship a srcset and let the browser choose.

Rule 2 — quality depends on what’s in the image

Photos: WebP at q80 — invisible difference.
UI screenshots with text: q90. At q80 the type goes mushy. This is the single most common mistake.

Rule 3 — a GIF is never the answer

GIF is a 1987 format: 256 colours per frame, no inter-frame compression. Same animation as H.264 video:

13.33 MB → 0.353 MB (−97%)
18.29 MB → 0.298 MB (−98%)

Same content, 40× smaller, better quality (real colour, no dithering), plus a poster so there’s no blank first frame, muted loop playsinline so it behaves like a GIF, prefers-reduced-motion support, and hardware decoding instead of pinning a CPU core.

Rule 4 — check the source before you blame the conversion

After converting, I compared the original with the WebP at 1400px and q90: the aliasing and the halo around the text were already in the original. So if an image looks bad, re-export it from the source file — don’t compress harder and hope.

The script

One shell script, one line per image with its role, and the original never gets touched:

images/work/arbor/arbor_cnp_2.png|figure
images/work/arbor.png|card
images/work/gfs/ss-no-filters.png|ui

Result on 95 files: 69.21 MB → 7.96 MB (−88%). No visible loss, because the widths and qualities now come from the layout.

The bug that cost me an hour

ffmpeg reads stdin. Inside a while read line loop over my image list, it swallows the rest of the list — so the file on the next line silently gets skipped. My script printed a mangled path and I spent twenty minutes convinced two files didn’t exist.

The fix is one flag: -nostdin.

None of this needs a framework with an image CDN. It needs knowing your own layout, four rules, and a script you run once. Which is the same reason I still build this site by hand: when you own the stack, the fix is a file in your repo, not a plan upgrade.

(If you’re building your portfolio from scratch, start with How to build your portfolio.)

The Melting Pot

Sun, 10 May 2026 21:00:00 +0200

The Melting Pot (idea)

A marketing history of New York, not a design piece.

The seed: at the Ellis Island museum, the Italian immigrant poster — “I heard the streets of America were paved with gold; when I arrived they were not paved at all, and they expected me to pave them.” More than 200 registries of Neapolitan immigrants named De Rosa, just like me.

The tension: America was (until recently) a safe beacon for immigrants — but maybe it never really was. The Statue of Liberty’s promise vs. the reality of slavery and inequality. The luxury we see/see is a product of that inequality.

[!NOTE] Editorial brief Idea: NYC immigration/marketing history — the “paved with gold” myth vs. reality. Personal (200+ De Rosa registries at Ellis Island) + political. Research so far: Black Panther Party (Newton & Seale, 1966) grew from the North’s failed promise — jobs/suburbs fled, black population ghettoized, police nearly all white. Statue of Liberty (1886, “Give me your tired…” = Emma Lazarus, 1883) while slavery’s legacy persisted. More to research on request. Audience: human personal essay — NOT career content, but shows range + writing. 🟣 life·place. Status: idea only — was a mis-titled duplicate of the Cuarón essay; corrected.

Homogenization (is it bad?)

Sat, 04 Apr 2026 16:01:35 +0200

• Dieter Rams at Braun — the 60s line of radios, turntables and small appliances, with his famous “less but better”, is the obligatory reference point against today’s generic minimalism.

• Marco Zanuso and Richard Sapper — Italian appliance design (Brionvega, for example the Algol TV).

Contrast with today

Today the equivalent would be studios like Pininfarina (which now draws the Battista electric hypercar), or within the brands, very large in-house design teams (BMW, Hyundai) where the individual designer’s name almost disappears from the public story — in contrast to personal signatures like Giugiaro’s or Gandini’s.

[!NOTE] Editorial brief Idea: why do modern cars all look the same? → regulation + aerodynamics (and maybe that’s good). Link to retro nostalgia (games) + ethics (the 70s polluted/were dangerous = bad design). Sub-threads: brand consolidation (a shared art direction); trends; Jony Ive / myth of the solitary genius (genius or surrounded by people? iOS flaws?); why a Ford Mustang is “damn good” and a modern SUV isn’t. Research so far: homogenization = regulation (lights/emissions/pedestrians) + aerodynamics (“hostage to the wind”). Do not treat lightly — add the Wences Sanz framing (regulation as something good). Audience: critical thinking / taste. 🟠 homogen·craft. Link to high fidelity and retro gaming + High_Seriality_Prototyping.

Fidelity and the Retro

Sat, 04 Apr 2026 16:01:35 +0200

Fidelity and the retro (idea)

The fidelity of the medium we’re emulating: artists designed sprites knowing how they’d look on CRT screens (they may even have designed on them). Playing the original Sonic on a modern HD screen doesn’t look like what it was.

Metroid Fusion was designed as survival-horror/metroidvania, but the GBA had no backlight → they used very saturated colors to compensate. Today, emulated or on Switch Online, it doesn’t look like the “dark and creepy” game where the Sa-X chases you.

→ Links to prototyping in code instead of Figma, breakpoints, responsive design. And to the theme of designing in the final medium (comic artist, A5).

[!NOTE] Editorial brief Idea: CRT fidelity — old games were designed for the display they shipped on; emulation/hd destroys that intent. Sonic (CRT vs HD) + Metroid Fusion (saturated colors to compensate for no-backlit GBA). Bridge: prototyping in code (final medium) vs Figma; breakpoints; responsive. 🟠 homogen·craft. Status: idea only (was an empty stub titled “Retro”).

Campus 3D

Sat, 04 Apr 2026 16:01:35 +0200

Big gamer since 6. PS1, Game Boy…

Juegos viejos Empecé con MAME. Luego la PSP. Siempre me han acompañado (los juegos retro)

Histus.

2020 back with Switch. Naturally get a Steam Deck, the perfect emulation machine. And bear with me, cause I need to tell you all this. EmuDeck. Not so perfect (Metroid Fusion), DS? Emu on iPad (meh) Small games, small screens. Anbernic. Another Anbernic (fidelity, aspect ratios) Retroid Pocket for Ps2/small Deck. Finally: Steam Machine.

Y ahora qué hago con los roms y los saves?

Romm + Tender + Syncthing on raspi + tailscale + iisu

What’s next: pandora/anbernic box + descaler de señal.

From Akinator to Backlog: the first prototype

Sat, 04 Apr 2026 16:01:35 +0200

The problem

I own hundreds of games I’ll never play. I’m not the only one: the industry is designed for you to buy, accumulate, and move on to the next thing — not to enjoy what you already have. The tracking apps that exist don’t help much — they tell you about the next release, they recommend the new stuff, but they don’t answer the question I actually ask myself every night: what should I play today, out of everything I already own?

That question was the starting point, before Backlog as it is now even existed.

The first idea: an Akinator-style recommender

My first prototype wasn’t a backlog-management app. It was something much smaller: a recommender I installed directly on my Steam Deck.

The mechanic was inspired by Akinator: instead of giving you a list or asking you to filter by genre, it made you choose between two games from your own library. Each choice halved the rest of the options, until you reached a single game. No infinite scrolling, no analysis paralysis — just quick, binary decisions until you got an answer.

The problem with Steam’s categories

For those binary choices to make sense — so it wasn’t literally heads or tails — I needed to compare games by real characteristics: tone, pacing, length, mechanics. Steam’s tags aren’t enough for that.

The solution was to pull tags from an external video-game database and build my own categorization layer, richer than Steam’s, so the binary decision tree had real criteria behind it.

What didn’t work (and what stayed)

This prototype didn’t quite take off as a standalone product. But it wasn’t wasted time: the categorization layer I built then is, right now, the foundation I’m reusing in Backlog, the project I’m working on with my friend Carlos. The idea is to use those same tags, further down the line, to feed a kind of pet or Tamagotchi inside the app that recommends or comments on your library.

It’s funny how a “failed” prototype ends up being the foundational piece of something else.

What’s next

In the next post in this series I tell the story of how we went from this solitary recommender to Backlog: an app where you can connect your Steam library, triage your games Tinder-style (swipe yes, swipe no), and even — with a fairly creative workaround — import your Nintendo history from screenshots and an LLM.

This is the first post in a series where I document, in real time, how we’re building Backlog.

[!NOTE] Editorial brief Idea: anti-consumerism in games — enjoy what you own, not buy-buy-buy. From an Akinator-style Steam Deck recommender to Backlog (with Carlos). Series: part 1 of the Backlog series (series_part: 1). Next: Tinder-style triage + Nintendo history import. Audience: indie/product hustle; links to the backlog case study. 🟢 consumption. Note: there’s a mis-titled duplicate of this under campus 3d.md (cleaned separately).