Since launch, there has been a phrase that has been associated with the Trongate framework. We see this phrase coming up time and time again. It’s a phrase that I have used, repeatedly, when describing Trongate. It’s a phrase that has attracted a lot of controversy. It’s a phrase that is a source of pride for Trongate developers, but a source of irritation for PHP developers who have not yet switched to Trongate.
The phrase is ‘industry first’.
Previously, I told you:
- The mechanism by which the Trongate framework handles updates is an industry first.
- The HAVC architecture that Trongate uses is an industry first.
- The fact that Trongate is PHP’s first ‘truly modular’ framework is an industry first.
- The very fact that Trongate is the only framework that gives you a free graphical query builder is an industry first.
- The very fact that the Trongate Module Market gives the PHP community a credible alternative to Packagist is an industry first.
- Trongate, having it’s own JavaScript libraries and CSS libraries - that’s an industry first.
- Trongate being a framework that aims to be ‘v1 forever’ is an industry first.
The list goes on and on.
Today, I have another industry first to bring to the table.
That’s because as far as I can tell, 2023 is going to be a year like no other, for web developers. Let’s be clear - there is only one topic that counts this year. It’s a topic that overshadows the entire web development industry. It’s more than just game changing for PHP - it’s industry changing. I am speaking, of course, about the rise of AI and - specifically - the new Open AI platform.
There are currently lots of developers speculating about whether or not AI will bring about the end of web development as we know it. Personally speaking, I don’t have a crystal ball and I’m not interested in engaging with those kinds of discussions. As far as I can tell, the important thing - at this time - is to make a decision:
Are we going to fight against AI or are we going to embrace AI as a tool that can potentially make us, and our framework, even better?
For Team Trongate, the decision is clear. We choose to work
with AI.
Therefore, I can exclusively reveal that - for Trongate - the transition has already begun. Within the last week or so, we’ve started the process of integrating code that was written entirely by artificial intelligence, specifically, the Open AI platform. Moving forward, I’m predicting that more than 50% of code that gets written for the Trongate framework will be code that was generated entirely by artificial intelligence.
Shall we all say it together?
1… 2… 3…
It’s another industry first.
Of course, I think there are a few important caveats to add. The first being an acknowledgement that whilst Trongate may be the first framework to be written (at least in part) using AI, it certainly won’t be the last. Also, having now had a chance to spend some time working with Open AI, I think it’s important to recognise the fact that it does not, as yet, produce perfect code
on command.
To elaborate on this a little more, I think Open AI is currently the world’s best JavaScript programmer. On the JavaScript front, Open AI is astonishing. I give it ten out of ten. However, on the PHP front, it’s not quite as strong. I dare say, everybody who has contributed to the Trongate framework is better at PHP than Open AI. For example, if you ask Open AI to write a function that encrypts a string, it will (at the time of writing) give you something that’s ‘bog standard’ and not entirely in line with the recommendations of the security heavyweights at OWASP. So, it’s certainly far from perfect.
Nevertheless, there are three areas - specifically - where AI can have an enormously positive impact. They are:
- Security
- Documentation
- Module Writing
On the security front, I can confirm that Trongate has already started getting loaded with AI generated code to help with things like string sanitization. This is the kind of area where AI is very strong and PHP frameworks are traditionally weak. Below is an example of some AI generated code that is now part of the Trongate framework.
function filter_name($name, $allowed_chars=[]) {
//remove HTML & PHP tags
$name = strip_tags($name);
// Apply XSS filtering
$name = htmlspecialchars($name);
// Create a regex pattern that includes the allowed characters
$pattern = '/[^a-zA-Z0-9\s';
$pattern .= !empty($allowed_chars) ? '[' . implode('', $allowed_chars) . ']' : ']';
$pattern .= '/';
// Replace any characters that are not in the allowed list
$name = preg_replace($pattern, '', $name);
// Convert double spaces to single spaces
$name = preg_replace('/\s+/', ' ', $name);
// Trim leading and trailing white space
$name = trim($name);
return $name;
}
As you can see, Open AI is good at producing code that contains awkward elements, like regex patterns. This kind of code can often be problematic and error-prone when written by us,
normal human beings. By embracing the power of AI, we're confident that we can make the Trongate framework much more robust than ever before.
On the documentation front, I’ve ran some tests and can confirm that Open AI is quite brilliant at describing how code works. As a matter of fact, I think it’s better at describing how code works than the guy who made the framework! So, I’m certainly looking forward to seeing Trongate soon having perfect, beautiful, comprehensive and error-free documentation.
Finally, I think the biggest technical challenge for Trongate has always been the challenge of populating the Module Market. Our main competition, in this regard, is Packagist and Packagist is absolutely huge. Packagist gives PHP developers access to many hundreds of thousands of libraries. I do
not think the Trongate Module Market will ever have as many libraries as Packagist. However, I think Open AI can play an essential role in helping us to generate the many ‘standard’ packages that we desperately need to make the Module Market an undeniable, great alternative to Packagist.
Do you know what’s interesting?
You’ve just read about a PHP framework now going down a road that, by any normal standards, would be utterly groundbreaking.
However, I don’t think you’ll be hearing about this on PHP News. They, and others like them, would much rather tell you about yet another boring Laravel version or how there are more certificates for sale with Symfony. Perhaps they'll babble on about some conference that you can get into for a four figure price tag.
Let’s start the year the way that we finished last year - by talking honestly.
Trongate
continues to be actively ignored by the various influencers of the web development community - particularly on YouTube. They’re ignoring it because we - of the Trongate community - had the balls to stand up and say, ‘we disagree’. We rocked the boat by re-introducing ‘pure PHP’ to the PHP community. We boldly told the self-appointed aficionados that we disagree with; their frequent rewrites, their certification and their over-engineered bullschitt. Then, we released Trongate - a framework that’s twenty times faster than the two market leading PHP frameworks. They can't handle it. They don't want to be seen as taking sides with a bunch of renegades and rule-breakers. It's much easier for those influencers to play it safe and that means pretending that Trongate doesn't exist, and perhaps hoping it goes away!
Trongate has been actively ignored by people like Brad Traversy, Stef Mischook and many others. We, who believe in pure PHP, have been relegated to the obscure pages of GitHub for long enough. It has been like a prison sentence. For some of us, it has been financially devastating. At times, the attacks have been personal and even career-wrecking.
But we’re still here.
I tell you today that THIS is the year when Trongate becomes a top ten PHP framework, based on GitHub stars.
All of those people who have actively ignored and attacked Trongate cannot hurt us any more. They have played their hand and we have received our punishment. But here we are, with over 700 GitHub stars and every day we get stronger. We’re not going anywhere!
I want all of you to know, I will NEVER accept a universe where Trongate is some unknown obscure framework that you have to go through some list to find. We WILL have a top three PHP framework, one way or another. That's because web development
matters. Keeping the doors of web development open
matters. As for those influencers who have actively ignored us? This year they have a choice. They can either change their behavior immediately, give Trongate the recognition that it deserves and be welcomed joyfully - by all of us - in a spirit of friendship, forgiveness and redemption. Or, they can continue to ignore Trongate - in which case I will personally see to it that the ridicule - for them - shall echo for a decade. What they do NOW defines them. Nobody has to like Trongate. Nobody has to like me, as a person. All we're looking for is a thoughtful recognition of the fact that there
is an alternative way of doing PHP. We ask for nothing more.
May you have a healthy, peaceful and prosperous 2023. Happy new year!
David Connelly
