Memoirs of a greybeard dev (or how web dev changed since 2000)

I am old. I know this because there are people working now who were not born when I started building websites for money. So, in the time-honoured tradition of us old gits, I'm going to talk through the last 20-odd years and how the industry has changed in that time and some of the trends I've noticed. This'll be part therapy session for me, part history, part autobiography. I'll begin by telling a story of my career, and then talk about how the industry has changed over that time and what's good and bad about it.

My Career - The early years

So picture the scene. It's 2000, I've just made a complete hash of my HND in computing, and I'm young and stupid. I was building websites for fun, nothing hugely fancy or exciting, but it was enough of a portfolio to get myself a very entry level role at a company definitely paying below the legal wage (the national minimum wage having come about a couple of years earlier). I was on £45/week and having to supplement my income by working nights in a freezer warehouse. Tough times.

So that first job, I turned up having blagged a job as a designer. I quickly established that I was nowhere near as good as my fellow £45-a-week professionals. In the room we had a bunch of young chaps mostly doing exciting things with Flash, as was the fashion at the time. I needed a niche, and noticed that while we had design, and we had Flash, we had a gap and that gap was in back end code. So I went and learned PHP. Lots of trips to devshed and the PHP manual online and I eventually got some cool things working. A side order of WAP (not related to the song, thankfully) followed, but in the end the company folded (quite how they managed this when paying us so little is anyone's guess, but I imagine things like one client hovering behind one of our developers for a year telling him what to do was a factor).

Next job then. A little company, this time paying a legal wage that allowed me to afford the rent on my tiny bedsit. 2 devs, one designer, a couple of sales people and a secretary. They worked in Coldfusion, connecting Access databases to their Coldfusion sites, which meant that a very small number of people could connect to their work before it went bang. On arrival I spent a week reading Sams Teach Yourself Coldfusion in 24 Hours (a misleading title if ever there was one - we all need sleep), and then set about fixing their broken code left by the previous dev. Literally nothing worked. I eventually persuaded them of the merits of PHP and MySQL as a superior choice.

There were no frameworks in those days, and Wordpress was still a couple of years away from creation, and Drupal was a bit closer but not particularly useful when it did arrive. I was working in Wales, which necessitated typing in a lot of Welsh. So I did what every self-respecting dev in the early 00s did and built a content management system. My CMS was magnificent, and could cope with unlimited languages (so I could avoid the Welsh typing) and had a plug-in module system to add things like contact forms, search, and so on.

One of the joys of that time was that we were building sites for small businesses, or people who wanted to start small businesses. I remember making a gorgeous website for a local company selling second hand refurbished laptops - our designer did an amazing job on it. Would anyone bother now? They'd just sell them on ebay instead, and have a presence on Facebook and Twitter. Back then though the internet wasn't homogenised as it is today, and small companies had their own sites and it meant we could help people with their dreams. It felt great.

Eventually the company went bust and I went to work for a company in Pontypridd - bit bigger, bit more organised, but still a similar structure, still basically a small local business. More money, which was great, I could afford a bigger flat. My CMS tech helped us shift a lot more sites, while my colleagues built stuff in ASP classic. One day I came in to find all the ASP sites had been hacked and my PHP ones had not - I felt quite smug. I unfortunately became an arrogant git and departed the company because of that, and looking back I was a real dick head when I was young. Not long after, they went bust. Three in a row. The industry was unstable and work was getting hard to find. Time for a break from IT.

A short break.. teaching, rocket launchers and growing up

So I spent a few years out to grow up and become less of a dickhead, during which I taught at a local college, did a degree in psychology, ran a nightclub photography business, and taught kids maths and IT in Thailand before coming home because people were firing rocket launchers at trains. I headed back to the IT industry after realising that teaching would never get me a mortgage.

Back to the code

My skills were horribly out of date, but thankfully a company in Bristol took me on as a junior dev and I quickly upskilled learning jQuery and working with CodeIgniter, Wordpress and Drupal. The company was a bit more organised than before, with a project manager present, but his role seemed mainly to be to keep dragging people from thing x to thing y as thing y was suddenly the most important thing in the world. The atmosphere though was very much the fun atmosphere that I encountered earlier in my career - it hadn't fully corporatised yet.

On to my last PHP job, and here I saw trends that I didn't like, and still don't like. As with the previous job we had micromanagement making it hard to get things done, but also so much of our work was now Drupal, sticking lego blocks together and trying to wire them up in spaghetti code. I was on a lot more money but a lot less happy, bored with work that was samey and lacking challenge, and the direction of travel looked bad. Time for a change.

From PHP to Java

I posted a message on a mailing list (yes I'm old) asking for advice on how to get into Java, and a chap replied, we got talking, and I ended up working under him at a local recruitment company. I learned Java, the company was I think fairly atypical of the industry in that the dev side had a lot of autonomy and no real interference from management. We got done what the company needed done. I had a good time, learned a great deal, and bought a house.

Covid happened, I got burned out after losing my entire team, and I left. Off to a Swedish military software company. This was a far more structured organisation, though still with a good atmosphere, and the project managers were starting to understand how to do the job in a more helpful way. Unfortunately we couldn't sell anything in the UK so I decided to try my luck elsewhere at an insurance company.

Insurance is a highly regulated industry, and it was by far the most corporate experience of my life. Getting anything done required figuring out what ticket to send in what format to what person with no clues left anywhere as to how you might go about doing that - the left and right hands very rarely communicating, while documentation was often missing or contradictory as procedures changed over the years and sometimes got documented and sometimes didn't. I don't believe that large organisations work, humans don't scale brilliantly, and nothing about the experience changed my mind.

So I now find myself off to pastures new, a more agile tech firm that features smaller self-contained units, hopefully this time I'll have a bit more fun. I'm old now, I have a kid and a mortgage, and I'm a lot less angry than I was in my 20s. I'm ready to look after a team and have some fun.

TLDR - what changed?

So after all that waffle, the important bit. What changed in the industry? I can see a few clear trends.

Companies got more organised/professional

I had 3 companies go bust in a row at the start of my career, but as far as I know all the companies I've worked for the last few decade are still going. It was the wild west, we didn't know how much to charge for websites, we didn't know how to make a profit because clients would change their minds on things, we'd not charge for that, and the thing would go over budget and become unprofitable. Every time. Fixing that is clearly a good thing, and I think probably the extra layers of management have helped that to some degree, but probably also companies figuring out better relationships with their customers and becoming more focused on PnL.

Structure got more rigid

We now have product owners, scrum masters, project managers and a whole raft of assorted individuals telling devs what to do. It's a harder game than when I was young, and if I'm honest a bit less fun.

Barriers to entry

In 2000, if you had skills in PHP, MySQL, HTML, CSS and Javascript (and could make all that work in IE6 and Netscape), and just spent a bit of time building a few things to put in a portfolio, you could get a job, albeit one that paid peanuts. It would however get you on the ladder. Now, even entry level stuff has very high requirements in terms of knowing a range of frameworks and languages, and while there is more free educational material available online now, I think this presents a significant barrier to new people coming into the industry. It's MUCH harder now than it was when I started, and I think may be why the newer recruits are more middle class bringing degrees as a way to prove skill and intellect (though some of this also comes from recruitment focusing more on university education).

A lot more money

I got into the world of web development simply because I liked code. I'd been programming since I was a kid, back in the early 90s I'd get free programming tools on magazine cover floppy discs and build games on my Atari ST, and this seemed like a fun way to make enough money to hopefully pay the rent (initially it was paying a lot less than the rent). There wasn't really an obvious career path, not many people really wanted to get into the industry because there wasn't much money to be made.

These days of course things are different. You can earn very good money as a web developer, though of course to do so you need a lot more knowledge. As you learn more, you earn more, a meritocracy that isn't universally shared by other industries. That meritocracy is the single best thing in IT. Nobody cares if you're male, female, white, black, gay, straight, anything else. All that matters is what you can do. You're a list of skills with a meatbag attached - feed the meatbag coffee and pizza and software comes out. This is not a bad thing. If you can build cool stuff, you WILL do well in this industry.

The work carries a lot more prestige now and more people want to do it, lots of whom had little interest back when there was no money in this game. This brings its own problems and changes to the culture of development, I'm not sure all of it is good, but that's a talk for another day.

More collaborative and taking longer

Back in the day I could build solo, just taking a PSD from a designer and cracking on, getting code into the client's hands in a few weeks. Now, all the frameworks and tech designed to make things more efficient and faster seem to have led to things taking longer and needing more people. It feels like we punch way below our weight these days. With so much time taken in meetings, so much time battling with frameworks and various bits of tech around the edges, we get a lot less done.

Skill Inflation

This one is a bit of an extension of the barriers to entry in some ways, but it feels like us older devs have pulled up the drawbridge a bit, though perhaps not intentionally.

Devs wanted more money, and to get it they needed to prove certain skills, so they started putting tech X into project Y regardless of its utility to that project, to pad their CV. Then of course maintaining project Y now required tech X meaning that hiring for that project required higher skills. Now multiply that out thousands of times across the industry and we see so many projects built with tech that really isn't helping anyone, just making things more complicated than they need to be.

The impact of this is to raise barriers to entry, to make it all so expensive that small companies can't come to us, to make so much of what we build a complete pain to maintain, and to slow down creation of new software. And worse still, it's a loop that we can't break out of because people will always be ambitious as long as this industry has so much money pouring into it.

More corporate

In those early days I was helping people realise their dreams of starting a small business with a website, or use a website to enhance their existing small business. Not so much now - those small businesses often don't exist, or do their work via Amazon, eBay or social media, with less and less need for an e-commerce site. So what's left is the corporate world, which is probably why there's now so much more money in the industry now, but I can't help but feel we've lost some of the fun.

What Does The Future Hold?

I'd like to see some of these trends roll back, with barriers lowering to allow young working class people into the industry at scale, and costs reduced to allow us to serve smaller businesses. I'd love to see the internet become less of a homogenised mass, with more businesses taking charge of their own sales instead of doing it all via Amazon, eBay and social media. Will those things happen?

Final Thoughts

There was a time when I'd unreservedly recommend young people get into IT, especially if they were on the autism spectrum, for no other industry was better for them. Now, I feel a bit differently about it.

First, it's a hell of a good thing that as you learn more you earn more. It's the single best thing about the industry. If you've got the appetite to do it, go for it. Second, you won't be working nights in a freezer warehouse like I did. However you won't be able to move with the agility I was able to, and to experiment in smaller projects. You will need to learn a million frameworks and technologies that you can't just run on your own machine as you could back in the day.

Finally there's a moral question. Back in the day we didn't think much about it. We were building stuff, it was cool, and the internet was just a tool, about which most people were optimistic. This isn't the case today as it has become corporatised and we see the damage it has done to media with the rise of clickbait, and to teenagers with the rise of social media. It's a much less optimistic industry and one a lot more afraid of the damage it can do, and rightly so, and so we now temper our enthusiasm for the amazing tech out there such as AI with a realisation that humans will probably do something awful with it.

I'm not sure I'll encourage my boy into this life when he gets older, but I guess we'll just have to see what the next couple of decades bring - maybe it'll be better then?