Why I stopped creating PHP projects with Wappler and why you probably should too

That is not true at all. You actually need to load an apache module to run PHP.

Tom, that’s not really a proper comparison. You don’t need docker to run NodeJS :slight_smile:

@JonL, aside from websockets what can you do in Node.js that you can’t do in PHP in Wappler?

I’m not against Node, I’m just curious. :beers:

Feature-wise it is about the architecture and the paradigms on which it was built.

It is not so much about being able to do something that PHP can’t and more about how it does it.

That being said there are numerous articles in the web that explain the key differences.

If I go all technical I don’t think I will help you understand.

PHP can do sockets by the way. Sockets is not exclusive to nodejs.

You don’t actually need to to load Apache to run PHP, but I think you misunderstood my point. You were pointing out that docker etc. and program languages are quite independent - and of course you’re right. I was suggesting that Apache and PHP (for example) could also be viewed as completely independent in the same way, but they also work so seamlessly together (and with MySQL), that they can also be considered more or less as a unit (which makes using and supporting them easier). Perhaps it wasn’t a very good point anyway…

I realise that. I don’t think I suggested you did, but I probably wasn’t being very clear.

Node.js is only supported for expensive hostings, at least at our provider. This is a no-go for smaller companies (projects).

How should a slow migration be possible if Wappler 4.0 had no php support? I currently don’t know what this migration would mean, but we have a cms with over 2000 server actions based on php.

You’d force Wappler user to move to Node.js just because you think Node.js it the right way to go? Glad George is not thinking like you :wink:

4 Likes

14 Likes

This is not true nowdays.

I said limited support. No new features. First with bug fixes and then only security fixes. In a period of time that makes sense. I said remove php from the 4.0 branch so people don’t select it for new projects, but users with a subscription to be able to use 3.x legacy branch where PHP is still usable.

They can easily be migrated.

Without hesitating.

:man_shrugging:t3:

Apologies. I meant to say to serve php web pages in Apache.

There is no tighter integration between php and apache and mysql than there is with any other major language.

Actually for PHP production environments you need to load a full external module so that apache can interpret it in a performant way.

While with node you just set apache in proxy mode as nodejs comes with it’s own production server.

WAMP/LAMP/MAMP is not some kind of magical software as many of you think. It just bundles some software into one installer and then changes some configuration lines to make them all work together in not so much peace and harmony.

And nobody created Apache, PHP and mysql to work together.

Docker on the other way. That is pure magic :wink:

1 Like

@Mr.Rubi I was meaning to reply to you but I got dragged into a spiral of fear of change :slight_smile:

Caprover is great, but check out also some new PaaS Open Source software based on k8s, terraform, etc

There is quite some movement on this front.



@george also mentioned the possibility of integrating terraform in the future when the user base grows

The co-founder of porter is here if you have questions or want to test their service on their cloud account.

2 Likes

Just saying. Node is not new. Has been around for over ten years. If we (our company) are honest, and in hind-sight, we probably should have started our switch-over at least a year or two back. Now for most of our current Projects we are starting to create parallel Projects based on Node and migrating. Initially we found it a little complicated as with all new things (as in new to us). But as it stands having the choice to do this is a superb reflection of Wapplers abilities. I don’t want to appear as picking on you Antony but ten years? Just reflect on that for a moment… Imagine if your host had not updated in ten years. Or if service providers were providing speeds of a decade ago for your broad-band? Step back ten years and think about how much has changed in that time period. Its an age ago my friend.

2 Likes

You know what’s (double)ironic about this?

People here that are pushing back on javascript(and specially nodejs runtime) can do it thanks to Jeff Atwood. The author of that quote and the creator of the software that powers this community(Discourse).

But the joke is on him. Discourse is not fully js. Its backend is written in Ruby :joy:

For a large percentage of developers, and for many years, it has meant hosting was largely a straightforward matter, allowing developers to focus on things they were probably more interested in. I think that’s pretty magical. Perhaps it’s like the internal combustion engine - familiar, much loved, robust, and pretty magical - but not the future (but still a few years before it disappears)

The lack of peace and harmony may be apparent to you, but probably not to the owners of the sites or those who you use them (and they’re important too). I doubt if many of my customers know (or care) what PHP is, although they’ve probably heard of it. (They almost definitely won’t have heard of NodeJS.)

I think that would be very premature - and probably disastrous for Wappler and many of its users, and from the point of view of attracting new customers.

1 Like

Yeah. Some customers could be lost sure, but others(newer generations) would be attracted also, don’t you think?

I definitely don’t think it would be disastrous for Wappler. As they could keep 3.x legacy for some time.

What I don’t think is sustainable is them supporting 5 different backends(asp classic, asp.net, php, nodejs and go) and keep being a small, agile and community centric team.

Hi @JonL! I am grateful to you and I really appreciate your help. I have read the proposed solutions, but if I understand their essence correctly, they will not be able to help in my task, unlike CapRover.

If I understand correctly, Qovery and Porter are targeting cloud hosting like AWS, DO, Heroku. And cloud hosting is not suitable for me. I don’t like any solution that only works with a predefined set of hosting solutions. The reason for this is that my customers dictate to me exactly where the project should be placed. This may be an unknown hosting company of absolutely insignificant size, but my customer has been drinking beer with the owner of this hosting company every day for 10 years. Or it can be the customer’s own servers. In other words, with all my desire, I do not have the opportunity to look only in the direction of cloud providers. I need to be able to simply and quickly place the project anywhere.

CapRover on the contrary solves this problem. It can turn any machine into a convenient hosting service, where you can host any projects.

It remains only to figure out how to most conveniently host Wappler applications using CapRover. I’m having a hitch with exactly that at the moment.

P. S. By The Way. For everyone who reads this thread. CapRover is a tool that makes life easier not only for nodejs developers. This is a piece that makes it easier to host and manage applications created on any technology stack. These can be your PHP or ASP projects. Or even stacks that are not supported by Wappler, such as Python or Ruby projects. Therefore, it can be extremely useful to absolutely all of you, even those who now do not understand how CapRover can be useful to him.

1 Like

Indeed. Caprover will suit better your needs then. I haven’t got my hands dirty with it but I am planning on doing it very soon as I am testing Oracle always free cloud plan.

I am planning on deploying to Oracle free plan a meilisearch instance to power one of my app’s search engine.

1 Like

Perhaps they would, but would they necessarily be deterred by the PHP option too? They might have ‘legacy’ sites to maintain, using PHP. However, as I said, I think it would be premature. I imagine there is a gradual exodus from Dreamweaver to Wappler, apart from the huge number of PHP developers generally. That would probably largely stop.

I agree about asp classic (but I don’t use it). ‘Go’? Perhaps you know something I don’t. Another factor which I expect would be relevant is Dreamweaver. I have no idea about the relative proportion of users of DMX’s DW extensions to Wappler users, but it might be signficant. Maintaining all this must be a lot of work, but presumably the PHP extension for DW and Wappler are more or less developed in parallel. To pull the plug on their DW users might not be a good option. I expect Adobe might pull the plug on DW first.

2 Likes

There seems to be a correlation between the decay of PHP in popularity, the rise of JS and the median age of the developers. You can check the data in stackoverflow’s annual surveys.

1 Like

To agree to disagree…

… to the point that hosting that supports Node JS is to expensive… yes sure if you have “bigger” clients footing the bill for bigger projects… but in my case i have loads of small clients that uses shared hosting… its really cost effective and no need for expensive hosting… im sure there are lots of Wapplerians here that have many smaller projects on shared hosting…

Just here in South Africa you can get hosting for R 70 - R 100 per month… shared hosting with all the php you want…
Then the JUMP from there is VPS or Dedicated hosting… then you start looking at R 1500 and up…

So in the end you have 2 Wapplerians here…
Hard Core Nodjs guys with serious big scalable projects with clients without a budget limit…
And then the normal entrepreneurial Wapplerian… that works for himself or a small team that runs tons of smaller projects…

SO for now there will always be a place for PHP here … or i would hope so…

And I would love to know the ratio of PHP vs NODEJS users here on Wappler…

4 Likes