Philosophically, there's even less justification.
So Trongate gets rid of the complexity of a single if() statement, and introduces new modules and documentation to return the same functionality that the single if() statement contained.
Okay.... I'm sure that makes perfect sense.
3 days ago
#21
3 days ago
#22
“AI moderation returned an invalid response” — that’s not a particularly helpful response. We’re all techies here. Whelp, here goes another finit attempt.
I’m genuinely curious, What are some packages installed via composer that’s accessed via /vendor in the first place ?
My gut feeling is saying >90% of actually maintained packages are relying on vendor/autoload.php to be required in where it’s consumed.
So, is there actually a real world usecase in the first place?
From my experiences, where I’ve reached for composer has been in individual modules for stuff I’d rather not build myself, like office document parsing and generation.
I’m genuinely curious, What are some packages installed via composer that’s accessed via /vendor in the first place ?
My gut feeling is saying >90% of actually maintained packages are relying on vendor/autoload.php to be required in where it’s consumed.
So, is there actually a real world usecase in the first place?
From my experiences, where I’ve reached for composer has been in individual modules for stuff I’d rather not build myself, like office document parsing and generation.
3 days ago
#23
Just a few points about the AI features:
1). I have removed the annoying requirement that forced everyone (including me) to constantly declare that something was for a personal project. It was easily the worst feature in the history of ...features.
Things should now feel much smoother.
2). When Grady submitted that post and agreed with me so emphatically, I can assure you that I was not cheating or engaging in any kind of manipulation. I built and launched the system during a live stream on YouTube today, so everything is there to see. I was as surprised by that post as anyone.
3). There are definitely some glitches with the forum, but they are being addressed. Within about seven days, things should be rock solid. Please bear with us and know that we are improving every day.
Regarding the topic of this thread, it feels as though two camps are forming: the purists and the generalists. I am firmly in the purist camp. However, regardless of where you stand, I want to assure you that I will not misrepresent or oversimplify anyone’s carefully considered arguments.
That is all I have. Anything further from my end would be repetitive.
Once again, thank you for your patience with the temporary AI experiments and the various forum issues.
DC
1). I have removed the annoying requirement that forced everyone (including me) to constantly declare that something was for a personal project. It was easily the worst feature in the history of ...features.
Things should now feel much smoother.
2). When Grady submitted that post and agreed with me so emphatically, I can assure you that I was not cheating or engaging in any kind of manipulation. I built and launched the system during a live stream on YouTube today, so everything is there to see. I was as surprised by that post as anyone.
3). There are definitely some glitches with the forum, but they are being addressed. Within about seven days, things should be rock solid. Please bear with us and know that we are improving every day.
Regarding the topic of this thread, it feels as though two camps are forming: the purists and the generalists. I am firmly in the purist camp. However, regardless of where you stand, I want to assure you that I will not misrepresent or oversimplify anyone’s carefully considered arguments.
That is all I have. Anything further from my end would be repetitive.
Once again, thank you for your patience with the temporary AI experiments and the various forum issues.
DC
3 days ago
#24
Playing around with the idea of packages in a module.
Here is the ai response.
Here is the ai response.
2 days ago
#25
Dan, you are an incredible person, and only you could come up with such an innovative idea!
However, the issue here has nothing to do with technical difficulties. It is more about making a statement of intent.
In the past, I have argued that Packagist is slow, insecure, fragile, and somewhat cumbersome. I do not expect everyone to agree with that assessment. However, a quick check on Google News does lend weight to my concerns regarding security. Furthermore, benchmark results also support what I am saying.
Now I am taking things a step further. I am suggesting that Packagist may have had its day. This is not just about Trongate. I am talking about the wider Packagist ecosystem as an outdated, pre-AI technology. The real disruptor here is not Trongate, but AI.
Personally, I have already started creating my own 'packages', and I have found the process to be both straightforward and easy. As AI continues to improve and more developers explore this approach, writing your own libraries with AI assistance will become the norm. This will not happen overnight, but I believe we are approaching a point where PHP developers begin to ask, “Do I really need Packagist?”
Trongate has an image problem. It's a “bare metals“ framework, and because of that some people erroneously think that Trongate is “old school“. In a world where thousands of PHP frameworks are competing for attention the idea of doing the kind of bold gesture that I've discussed in this thread would surely be electrifying. It would send a powerful message - telling everyone that we can see which way the wind is blowing and we are boldly looking towards the future.
I don't think there's anyone in the PHP space having this conversation. If we were the first to throw down the gauntlet, it would indeed be electrifying. It may deter some people from using Trongate. That's true. However, in my opinion, those people would be like the ActionScript developers in 2007 who refused to buy an iPhone because Apple doesn't support Flash.
That being said, I have failed to convince everyone in this instance. The failure is all mine. The Trongate framework is not some dictatorship. We have to respect other opinions - especially if they come from people who have stood by the framework for years.
So, all is well. There is no disagreement that needs to escalate, and we can draw a line under this topic for now. Perhaps we can revisit it in the future. It would be interesting to see how opinions evolve.
DC
However, the issue here has nothing to do with technical difficulties. It is more about making a statement of intent.
In the past, I have argued that Packagist is slow, insecure, fragile, and somewhat cumbersome. I do not expect everyone to agree with that assessment. However, a quick check on Google News does lend weight to my concerns regarding security. Furthermore, benchmark results also support what I am saying.
Now I am taking things a step further. I am suggesting that Packagist may have had its day. This is not just about Trongate. I am talking about the wider Packagist ecosystem as an outdated, pre-AI technology. The real disruptor here is not Trongate, but AI.
Personally, I have already started creating my own 'packages', and I have found the process to be both straightforward and easy. As AI continues to improve and more developers explore this approach, writing your own libraries with AI assistance will become the norm. This will not happen overnight, but I believe we are approaching a point where PHP developers begin to ask, “Do I really need Packagist?”
Trongate has an image problem. It's a “bare metals“ framework, and because of that some people erroneously think that Trongate is “old school“. In a world where thousands of PHP frameworks are competing for attention the idea of doing the kind of bold gesture that I've discussed in this thread would surely be electrifying. It would send a powerful message - telling everyone that we can see which way the wind is blowing and we are boldly looking towards the future.
I don't think there's anyone in the PHP space having this conversation. If we were the first to throw down the gauntlet, it would indeed be electrifying. It may deter some people from using Trongate. That's true. However, in my opinion, those people would be like the ActionScript developers in 2007 who refused to buy an iPhone because Apple doesn't support Flash.
That being said, I have failed to convince everyone in this instance. The failure is all mine. The Trongate framework is not some dictatorship. We have to respect other opinions - especially if they come from people who have stood by the framework for years.
So, all is well. There is no disagreement that needs to escalate, and we can draw a line under this topic for now. Perhaps we can revisit it in the future. It would be interesting to see how opinions evolve.
DC