Why php?

[CompiledMonkey]

Also, im wondering why you are interested?

So many people use it, I'm curious to hear their reasons.

Are you not satisfied with the technologies you claim to know?

I'm very satisfied with what I know. I'm looking at some of my future contracts and they are much smaller than what I'm used to. So I was thinking PHP may be a good choice because cheap hosting is there and the technology itself is free.

Are you bored?

School's over for the summer, in a way yes, I am bored.

Trying to improve your horizon (dought that, cause otherwise you would have taken a look at PHP first)?

Eh, not really. Just be prepared for any future requests or needs.

Not that i care personnaly, but i'm having a hard time to believe you SERIOUSLY wanted to evaluate PHP's capabilities.

Meh, ok.

[Jeremy W.]

Of course, eh Chris? Because anyone who doesn't believe PHP is the best doesn't seriously want to evaluate it I'd think PHP would benefit from someone who picks it out of critical examination

J

[CompiledMonkey]

Hey man, I need your expert advice! Check the .NET forum.

[freddydoesphp]

As far as freddy's quick "boil it down", those are the exact reasons I chose .NET, to be honest

If .NET is the easiest most cost efficient method for developing dynamic content then by all means you should stick with using it. Others find the other way around is better, but both are successful.

Use whatever you feel the best tool is to do the job. Thats what it boils down to.

[Jeremy W.]

Yup, I'm just saying that it isn't as clearcut as some folk would have you believe. I honestly believe PHP is better for smaller sites which aren't tied into larger systems. In fact, for every independent small site I do, I do it in PHP now

[Dr Livingston]

Fine. Whatever. Just don't give PHP a bad name.

That's all I've got to say really, okay ?

[CompiledMonkey]

Fine. Whatever. Just don't give PHP a bad name.

That's what we're trying to say about .NET.

[M. Johansson]

Why PHP,

Of course because its the simplest, cheapest way to get dynamic content into a website. Thats it!

I just wanted to quote this for emphasis.

[freddydoesphp]

Hey I sue .NET when I need to. Working with it right now. But the key is to know which one to use for the job.

I just wanted to quote this for emphasis.

I'll take it you agree. M. Johansson

[M. Johansson]

Hey I sue .NET when I need to. Working with it right now. But the key is to know which one to use for the job.



I'll take it you agree. M. Johansson

Aye. As rule of thumb, I use it when I'm on a miniscule budget, or if I'm building a forum (NOTHING beats vBulletin).

[mbb]

Aye. As rule of thumb, I use it when I'm on a miniscule budget, or if I'm building a forum

So when your customers have no money you offer them PHP as the solution? And when they have money to burn you rip them off with .NET?

I’m not sure if I understand this statement can you give me examples of when your budget has influenced your technology bias. The way I see it if customer A has £2000 to spend and customer B has £200000 the technology decision still comes down to their technical requirements and not how much they can afford.

I have probably misread some of the previous posts; but I’m sure I was lead to believe that all of this Microsoft technology was free now or that they paid you! Either way this is surely a reason not to include financial prejudices with your decision. Also it would be nice to see examples of when PHP couldn’t cut the mustard and .NET was needed to save the day, this would be helpful from an educational point of view as most of my work is done with PHP.

[freddydoesphp]

So when your customers have no money you offer them PHP as the solution? And when they have money to burn you rip them off with .NET?

I’m not sure if I understand this statement can you give me examples of when your budget has influenced your technology bias. The way I see it if customer A has £2000 to spend and customer B has £200000 the technology decision still comes down to their technical requirements and not how much they can afford.

I have probably misread some of the previous posts; but I’m sure I was lead to believe that all of this Microsoft technology was almost free now of that they pay you! Either way this is surely a reason not to include financial prejudices with your decision. Also it would be nice to see examples of when PHP couldn’t cut the mustard and .NET was needed to save the day, this would be helpful from an educational point of view as most of my work is done with PHP.

Very good point. I was just about to post that I feel project scope should be the determining factor. If a client has tons of money to spend, but all they need is some dynamic data on web page, why charge them double to use .NET when php will suffice. You'd probably end up keeping their business by cutting them a break. However on the flipside if a client has a low budget and wants something that the .NET framework as a whole would handle better, wouldn't it be better to explain this and let them know their current budget constraints don't allow for this and they need to rethink things?

[M. Johansson]

I’m not sure if I understand this statement can you give me examples of when your budget has influenced your technology bias. The way I see it if customer A has £2000 to spend and customer B has £200000 the technology decision still comes down to their technical requirements and not how much they can afford.

It's not really all that complicated. If a client does not need/feel that he needs a complex web site, he will give me a small budget to work with, and he will get a simple dynamic page, and that is more cost-effective to do with PHP, as .NET or Java would be overkill.

If a client has a larger budget for her web site (since she needs a bigger, more complicated site for her business) I would build the site in .NET or Java, because those two are better adapted for large projects than PHP, and therefore more cost effective.

I have probably misread some of the previous posts; but I’m sure I was lead to believe that all of this Microsoft technology was free now or that they paid you! Either way this is surely a reason not to include financial prejudices with your decision.

Yes, you have misread the previous posts, or they were miswritten. Developing .NET is free. All the needed development tools are provided by Microsoft, including Web- and SQL Server. Hosting, on the other hand, is more expensive than PHP hosting, especially if you need SQL Server, and that is a tradeoff you will have to deal with. It's a matter of what you need.

Also it would be nice to see examples of when PHP couldn’t cut the mustard and .NET was needed to save the day, this would be helpful from an educational point of view as most of my work is done with PHP.

It has never been what is able to "cut the mustard", in my opinion. It's what is most well adapted to do the job. Java and .NET are more adapted to large projects than PHP is. Out of the box, PHP offers extremely crappy separation of tiers, for instance. You can do very nice separation of tiers with PHP, but then you have to develop your own system for that, which is [b]extremely[b] costly compared to just using Java or .NET. Same thing with mySQL vs. MSSQL server. mySQL lacks a few vital features that MSSQL has. You can work around them with PHP, but that once again (mostly) costs more than it would to just use MSSQL Server. If you don't need tier separation (if your project is not large enough to benefit from it) or if you don't need advanced SQL features, you may be better off with PHP.

If you already have developed an existing PHP framework for tier architechture that can rival JAVA/.NET, and have an alternative SQL server at your disposal, and can sell them to the client for a cost lower than the Microsoft equivalent, by all means, do so!

[archigamer]

I hate to bring this thread back to life but one of the things that I find lacking in PHP is the ability to have constant variables survive longer than that page. take a look at OSCommerce. Possibly the finest free shopping cart on the net. looking at the setup there are over a dozen constants every page all being redifined every request. also not to mention there are dozens of constants in the database that are redefined every request. Also all the constants for that users langauge are requested along with the constants for that particular page. thus i am guessing there are about 100-150 contants created every single request to any page in the system. also not to mention the string connections made. If this could only be defined once and live throughout the scripts there would be a dramatic speed boost.

Also seperation of tiers is a pain which i consider essential. with .net I have implemented a CMP model (the same recommended by EJB 2.0) along with some XML metadata. The xml is processed once and added to an application variable. Yes CMP could be implemented into PHP but you would have to set up the CMP groundwork every single page request to any page. thus creating a slowdown.

[firepages]

If you don't need tier separation (if your project is not large enough to benefit from it) or if you don't need advanced SQL features, you may be better off with PHP

What has advanced SQL features got to do with PHP ? ... nothing , Oracle, MSSQL ... whatever , you know the list.

whatever did people do before .NET one wonders , must have all used JAVA I suppose because by reading this thread it is clear that only J2EE && .NET can overcome load balancing and connection pooling issues to deal with multi-tier environments ....

that the .NET and j2EE platforms provide such is surely true, that its the only way is just as surely false.

or perhaps because there is not a load_balance() command in PHP people think thats it

[firepages]

is the ability to have constant variables survive longer than that page.

variable variables, yes... but I have to admit that PHP has not yet grasped the concept of variable constants

Applications like OSCommerce have to , by their very nature be highly configurable, they have to take into account all the unknowns of the end-users system and setup etc, all reusable applications suffer from this.

If you were writing a shopping cart for yourself then you would not write it as per OSCommerce (unless you wanted to distribute it)

You can of course define PHP constants in the server configuration ... for a larger project you can even recompile PHP to include your constants ... how realistic this is depends on the scope of your application, the fact that you can recompile PHP or write your own modules to cover application specific needs is a gigantic bonus rarely appreciated (many of the existing PHP modules have been donated by this very system)

Looking purely at PHP scripts , or at the basic functionality standard with PHP is not seeing the whole picture.

[HarryF]

Just picking up on minor detail.

one of the things that I find lacking in PHP is the ability to have constant variables survive longer than that page.

This can be done with shared memory: http://www.php.net/manual/en/ref.sem.php FYI but it's not common practice (given the dangers) - there's a good article though here if you're interested: http://www.zez.org/article/articleview/46/. There are other ways to solve this kind of problem as well such as PHP Shelve

or perhaps because there is not a load_balance() command in PHP people think thats it

Well it does, if you're talking about balancing sessions across servers: http://www.php.net/manual/en/ref.msession.php

The bottom line as far as companies are concerned is the risk involved in finding the right developer. With PHP it's essentially possible (and often easy) to accomplish anything you might do with J2EE or .NET, in terms of end result but you will need an experienced developer. Yahoo, for example, hired Rasmus Lerdorf... (PHP's creator). With .NET / J2EE the risk (assuming basic competence) is probably lower as the frameworks enforce some standards.

As to the overall discussion and CompiledMonkey's question - my own preference for PHP is most of all because it's a dynamically typed language - I like the flexibility that brings from many angles - my attention span is probably too short and I get impatient with strongly typed languages quickly (none of which I'm particularily good at as a result). I also don't like working with code generation tools for the most part, other than UML tools and prefer the humble text editor.

I don't know if you've tried playing with Jython, which is a Python implementation in Java. Think I saw somewhere you've played with Java / Swing - you might be interested in this and this for an idea of how a loosely typed language can be used to achieve the same results.

For me that's what PHP does for building web apps.

[webmasts]

I think it is easier than other lanuguages for newbies. I tried learning PERL..it seems very complicated!

[M. Johansson]

What has advanced SQL features got to do with PHP ? ... nothing , Oracle, MSSQL ... whatever , you know the list.

It doesn't. I was merely making a point about choosing between cheap and expensive software.

[M. Johansson]

That was a truly fantastic post, Harry - here comes reputation!

The bottom line as far as companies are concerned is the risk involved in finding the right developer. With PHP it's essentially possible (and often easy) to accomplish anything you might do with J2EE or .NET, in terms of end result but you will need an experienced developer. Yahoo, for example, hired Rasmus Lerdorf... (PHP's creator). With .NET / J2EE the risk (assuming basic competence) is probably lower as the frameworks enforce some standards.

Precisely! For PHP to be an effective enterprise solution, you need a developer who nows what she's doing. You also need to use quite a bit of third-party software or write a lot of own stuff. Either way, it's quite a job to manage, and not for the faint of heart. If should logically be cheaper if you have the enourmous amounts of servers that Yahoo has, though.