Delphi For PHP Released

Gramie2 writes "Codegear (now a subsidiary of Borland) has just released version 1.0 of Delphi for PHP, a RAD development environment (running on Windows) that produces standard PHP code. It features a large set of built-in components, including ones that use AJAX for database access; and Codegear is encouraging users to develop their own components. The framework, VCL for PHP, is open source, and documentation follows the PHP model. Initial database connectivity is for MySQL and Interbase (Codegear's commercial database that spawned the open-source Firebird), but more are promised."

Comments

I for one

[Mipoti Gusundar]

I for one am welcoming or new component drived rapid development overloads. In an agile way!

Slashvertisement

[Anonymous Coward]

Damn Slashvertisements. Who gives a crap?

Finally!

[Anonymous Coward]

Finally - Delphi will give PHP the credibility it needs in the corporate world (hahahahahaha)

Disambiguation

[Zouden]

If you're as confused as I was, it's because the name Delphi can apply to the language Object Pascal, as well as for the IDE used primarily for Object Pascal.

This article is about the IDE being used for PHP, so fans of Pascal syntax have nothing to get excited about.

Re:Disambiguation

[PalmKiller]

Oh I was confused back in the day. I was using borland pascal, pascal for windows, in addition to c++. They send me a marketing flyer for delphi hyped up as a new and unique language and offering me an upgrade of one of my current products, no mention whatsoever that it was pascal on it. I threw it away cause I had no idea it was pascal...later I figured it out (Their sales told me no new pascal would be offered but they had a new delphi product that replaced it)...by then I had moved on to c++ as the only product of theirs I still used. I guess now I got the chance to get a back into pascal a little with the delphi/c++ combo in their new studio line, when and if I upgrade.

Re:Disambiguation

[mikepence]

I think that the thing to get excited about is the VCL for PHP -- stateful components in a web presentation layer, and CodeGears intense interest in Ruby and Rails.

I interviewed CodeGear's VP of Product Dev, Mike Swindell, about this product and their intent to get into the Rails market at http://mikepence.wordpress.com/2007/03/20/delphi-o n-rails/

Re:Disambiguation

[Viceroy Potatohead]

This article is about the IDE being used for PHP, so fans of Pascal syntax have nothing to get excited about.

As far as can tell, fans of Pascal syntax have nothing to get excited about, anyways... [Ducks]

Why does this eerily

[Anonymous Coward]

compare:

http://video.codegear.com/php/hello/hello.html

and
http://docs.rinet.ru/ActiveVB/f22-3.gif ...

Why does this IDE remind me of my visual basic 3 days (not to say that there's anything wrong with that). I guess I've been staring at eclipse for too long.

Re:Why does this eerily

[Anonymous Coward]

not to say that there's anything wrong with that

No, of course not.

Nice (so-called) dot-net alternative

[CarpetShark]

Personally, I think PHP is a pretty horrible way to do web development of any real size. Web apps need to be written in proper code, with MVC, and templates, not as code embedded in individual pages.

However, this might at least this might steer some windows users away from the thing microsoft likes to pretend is reponsible for a TLD ("dot net"), which is worse.

Re:Nice (so-called) dot-net alternative

[Anonymous Coward]

>> Web apps need to be written in proper code, with MVC, and templates, not as code embedded in individual pages.

What, you mean like every developer worth their wage in PHP has been doing for the last x years?

Do try and keep up...

Re:Nice (so-called) dot-net alternative

[Anonymous Coward]

You can use PHP to write web apps with proper code, MVC and templates through the use of frameworks (CakePHP, php.MVC, Zend, etc, or your own). The purpose of a language is not to provide you with all those things. You have to work for them, no matter what language you're writing in.

Re:Nice (so-called) dot-net alternative

[AutopsyReport]

Your way of thinking is outdated. There are many frameworks that facilitate proper MVC, templating, and a separation of duties. Check out CodeIgniter, CakePHP, and symfony, three of the most popular frameworks for PHP. Development techniques and approaches in PHP have changed significantly in the last few years.

Re:Nice (so-called) dot-net alternative

[tolan-b]

The embedding code in HTML part of PHP is fairly legacy really for anyone doing anything more complex than basic scripts. Any decent larger PHP app will only use that method as part of a templating/view layer (as long as there's no business logic involved it's a useful templating tool), if at all.

Our CMS is MVC (command and controller j2ee pattern specifically), using PHP's embedding in the views only, is fully OO and has an O/RM layer for datastore access. This kind of set-up is increasingly common in PHP now. Just look at the number of application and database frameworks available for it.

Of course there are always going to be kids knocking out horrible scripts, but that doesn't mean there aren't people doing things properly too.

Re:Nice (so-called) dot-net alternative

[sherriw]

"Web apps need to be written in proper code, with MVC, and templates, not as code embedded in individual pages."

No kidding? Have you ever seen a quality application written in PHP- it can do all these things and more. I've written many quality PHP applications that use a modified MVC architecture and has all the PHP code separate from the output templates. On top of that it uses OO where it makes sense to do so, it's fast and secure.

It helps to know what you're talking about before you spout off. Just because lots of people build rickety shacks out of stone, doesn't mean you can't build a solid castle out of it too.

It's this kind of generalization that pisses me off. It's not the language, it's your crappy skills.

Re:Nice (so-called) dot-net alternative

[boaddrink]

If this year was 2000 then I would agree with you. PHP has quite a few MVC frameworks that have been around for some time and are fairly extensively used in enterprise environments. If you know any developers embedding their PHP in the HTML in large apps please ask them to stop.

Here is a small example of a few MVCs out there.
CakePHP - http://www.cakephp.org/
Symfony - http://www.symfony-project.com/
Zend Framework - http://framework.zend.com/

Re:Nice (so-called) dot-net alternative

[Anonymous Coward]

Personally, I think PHP is a pretty horrible way to do web development of any real size. Web apps need to be written in proper code, with MVC, and templates, not as code embedded in individual pages.

Clearly you have never participated in serious php application development. It is always easy to criticise that which you have not experienced and do not understand. So do us all a favour and STFU.

And this totally ignorant post was moderated +3 interesting. What a bunch of clueless dipshits, to be duped by such blatant ass kissing.

But by all means, don't let me interrupt the ongoing predictable php bashfest, and slashdot perl ass kissing. It is going to go on as usual. Mod this post a troll, you know you want to the truth always hurts.

In the hands of an experienced programmer the performance of C/C++ or php for web applications is miles ahead of the ugly slug known as perl, and ruby is popular becsaue people like to say 'rails', the same people like to say ACID 2 and think its actually important.

But please, if you are going to shit on php, at least get a fucking clue.

Re:Nice (so-called) dot-net alternative

[Rui Lopes]

So use CakePHP and your problems and issues with PHP will be solved...

Re:Nice (so-called) dot-net alternative

[Anonymous Coward]

Ok, let's try replying in a mature fashion now, mkay? Don't weaken our arguments against this guy with offensive language.

Anyway, what are you mentioning the Acid2 test for? Acid2 is a test for a browser's compliance with CSS 2. There is no correlation between Ruby users and CSS compliance folks. And yes, the Acid2 test is important. The original Acid test encouraged browser makers to support CSS 1, and it has done the same in regards to CSS 2.

And have you ever used Ruby on Rails to develop a site? Rails is one of the most mature frameworks in the open source sector. I use it because a) it's a blast to program in and b) the amount of testing it facilitates is terrific. Before you label me, I still use PHP quite often for quick projects and for the things that Rails can't do that PHP can do easily.

Re:Nice (so-called) dot-net alternative

[CastrTroy]

What I like about PHP is that it's flexible enough to do either kind of programming. Being an ASP.Net programmer, I sometimes wish it were just easier to mix the code and presentation when making a quick page. On large projects, it's good not to mix the two, but sometimes throwing together a simple page is easier if you can mix code and presentation.

Re:Nice (so-called) dot-net alternative

[daeg]

Anyone can make a language or framework insecure. PHP just makes it easier than most.

Are you subscribed to the Secunia security mailing list? A good 1/3 - 1/2 of them are flaws in PHP applications with widespread installation bases. That says something about the language, whether it is the language itself or the prominent users of the language, but I suspect it is a combination of both.

Re:Nice (so-called) dot-net alternative

[suv4x4]

No kidding? Have you ever seen a quality application written in PHP- it can do all these things and more. I've written many quality PHP applications that use a modified MVC architecture and has all the PHP code separate from the output templates.On top of that it uses OO where it makes sense to do so, it's fast and secure.

As a PHP developer, I agree with everything you said, except "fast", unless:

1. we ignore the speed of all the other platforms out there (python, perl, .net, java).

2. your requirements of "fast" are modest.

Truth is with more complex architectures and lots of OOP, PHP is really slow, even bytecode caching helps only so much.

PHP shines speed-wise exactly with the kind of "html-and-php-code" soup most pro developers despise. When Yahoo claim they use PHP, they in fact use it as a templating language in exactly this kind of "soup", their actual backend is C and Java.

This is why I'm really surprised at what CodeGear is trying to pull off here. As a developer of an in-house component based template engine myself, I know how painfully slow PHP becomes when you try to abstract some of your logic away in classes and so on. Various "PHP OOP" efforts like Zend's own framework or EZ Components prove my point as well.

Delphi's visual approach with VLC is just a huge bunch of abstraction. I can only imagine the kind of speed these PHP apps will have.. In fact you can pretty much say this effort is doomed from the very moment "PHP" got involved.

Re:Nice (so-called) dot-net alternative

[Yoozer]

Being an ASP.Net programmer, I sometimes wish it were just easier to mix the code and presentation when making a quick page.

Dump everything in a Response.Write - or make a single Label called "test" you use to dump strings in (use a stringbuilder, though, concatenation is pure inefficient filth for everything bigger than 3 lines) ;)

Re:Nice (so-called) dot-net alternative

[nuzak]

Also check out Prado for a component-oriented framework along the lines of JSF or ASP.NET (its design seems to be inspired more by the latter). I deeply loathe PHP, but Prado does look very nice.

Re:Nice (so-called) dot-net alternative

[Dragonslicer]

Are you subscribed to the Secunia security mailing list? A good 1/3 - 1/2 of them are flaws in PHP applications with widespread installation bases. That says something about the language

That it's easy enough to learn that complete idiots can write crappy code in it? A bad programmer will write bad code in any language that you put in front of them. PHP just happens to be one of the few languages that's simple enough that the aforementioned idiots can write code that works (for varying definitions of "works"). Programming languages can only do so much to promote security before the non-idiots complain that the language won't let them do something that they need to do.

Re:Nice (so-called) dot-net alternative

[alittlespice]

not to mention Zend's solution: The Zend Framework.

Re:Nice (so-called) dot-net alternative

[Anonymous Coward]

What are <% %> tags for?
<%if(Request["step"] != "1" && Request["step"] != "2") { %>
<!-- some HTML Code -->
<%}%>

Re:Nice (so-called) dot-net alternative

[shutdown -p now]

No kidding? Have you ever seen a quality application written in PHP- it can do all these things and more. I've written many quality PHP applications that use a modified MVC architecture and has all the PHP code separate from the output templates. On top of that it uses OO where it makes sense to do so, it's fast and secure.

The point is that there is nothing specific to PHP that facilitates this approach, and a lot (such as its template syntax) which distracts from it. If you're going the OO/MVC route, why not use Java? Or, if you really like dynamic typing, Python (and at least get Unicode working properly), or Ruby with all its cool language features?

Re:Nice (so-called) dot-net alternative

[a.d.trick]

The complaints aren't all bogus. PHP does have a number of painful points. Probably the one that's given me the most trouble is error handling. PHP has exceptions, but most of the error handling in the language doesn't use exception. Then there's fatal errors which aren't really possible to handle (you can use an output buffer handler, but that introduces other problems which may/may not be worth the effort).

PHP also has those crazy globals floating all over the place. Sure, you can implement your own MVC-like architecture, but why should you have to do the work?

Finally, the object model in PHP has a number of deficiencies (C++/Java programmer may not notice; but if you've ever learned Python, PHP will make you cry). Much of this has been fixed in PHP5, but there are still a good number of things you can't do (mostly because classes are not first-class objects).

That said, it still is my first choice in most situations due to it's massive level of support from web hosts and acceptance among web developers.

Re:Nice (so-called) dot-net alternative

[nuzak]

<%if(Request["step"] != "1" && Request["step"] != "2") { %>
  <!-- some HTML Code -->
  <%}%>

That, fellas, is why you separate the logic out into templates. Any "language" that makes me have constructs like <%}%> needs to have a stake driven through its black heart.

Re:Nice (so-called) dot-net alternative

[sootman]

Yeah, and everyone should eat their vegetables, brush their teeth, drive the speed limit, etc etc etc. My code would probably make a seasoned developer weep but my apps are simple, they work, I'm able to maintain them easily, and hundreds of people use them daily with no ill effects. The whole point of PHP--and I've heard Rasmus Lerdorf say this personally--is to make it easy for people to crank out apps that are useful. Period. I don't stay awake worrying that some CS grad (or worse, some self-taught RoR fanboy) won't like that I've mixed data and presentation. I try to keep things somewhat neat, but if I need to jump into the middle of a table and say
<td<?php if ('somethingBad'==$currentValue) { print ' bgcolor="red"'; } ?>>
I will.

Re:Nice (so-called) dot-net alternative

[1110110001]

PHP also has those crazy globals floating all over the place.

Which crazy globals do you mean? You don't have any globals in functions or method unless you import them. Or do you mean the 9 super globals? I don't know what's crazy about them. The way they are written is very easy to distinguish from other variables. Thus if you don't want them to be used directly it's very easy to enforce your rules with a simple grep.

Re:Nice (so-called) dot-net alternative

[CarpetShark]

Admittedly, I'd never heard of CakePHP before. It looks like an interesting project.

As far as CakePHP solving the issue goes though... no, it won't. CakePHP is an MVC inside an interpreted-every-time V, which is part of a rebuilt-every-time MVC. This is very inefficient, obviously. It's pretty much doomed to be that way, because, as I said, PHP is designed to run inside pages, rather than to run and provide pages. CakePHP would be a reasonable alternative for MVC on legacy shared hosting that didn't provide anything but PHP, but for all intents and purposes, it should be avoided, imho.

AJAX for database access???

[Anonymous Coward]

AJAX for database access? WTF? I thought AJAX was between browser and server.

Re:AJAX for database access???

[CarpetShark]

They're referring to dynamically updating forms, such as when you have two combo boxes, select something in one, and the options change in another, or when you enter a search filter, and the table of found records updates without reloading the entire page.

Re:AJAX for database access???

[Gramie2]

I guess I could have been clearer. Controls on the page can be linked to datasets on the server. The mechanism is AJAX from the browser to the server; then the server deals with the database; and a return to the browser. But all you have to do is define a dataset, and set properties on the control (list box, table, etc.) to point to the dataset.

Re:AJAX for database access???

[./]

agreed, this was just buzzwords being flung with mad abandon.

Re:AJAX for database access???

[suv4x4]

I guess I could have been clearer. Controls on the page can be linked to datasets on the server. The mechanism is AJAX from the browser to the server; then the server deals with the database; and a return to the browser. But all you have to do is define a dataset, and set properties on the control (list box, table, etc.) to point to the dataset.

That sounds more like a security disaster. Freely linking stuff to the data source (including modification) is nice for a desktop app, but when the data is on your server, and the client (web page) falls into the wrong hands...

ASP.NET mitigates this my pushing the actual processing of data changes into the controller (they call it "code-behind"), but if Delphi for PHP blindly copies the data binding from Delphi... get ready for some hard-core hacking action.

Does this IDE build upon the existing interpreter?

[Anonymous Coward]

Does this IDE just use the normal PHP interpreter available from http://www.php.net/ or did Borland/CodeGear write their own PHP implementation?

Borland has died after Borland Delphi 7

[vivaoporto]

When Borland (then Inprise, then Borland again, then Codegear(?) ) stopped making sober RADs and decided to take a chance on expensive toys for code management, they lost in both fronts. The Turbo Series (Pascal, C and Assembler) and Delphi (the odd versions, 1, 3, 5 and 7) seriously competed against Microsoft products (Microsoft C, Assembler, Visual Series), even outselling them in a lot of places in the world (Brazil, for instance).

Two things made Borland wreck their scene: 1) losing their creative minds to Microsoft, specially Anders Hejlsberg, creator of nothing less than Turbo Pascal, Delphi and main architect of C#. 2) losing their focus (from useful RADs to expensive but totally good for nothing "Application Lifecycle Management" (whatever it is).

Had kept the focus and the creative minds, either .Net would not exist (and consequently, stole Borland's thunder) or the Borland tools would be better even than the Microsoft ones on that fronts (Delphi 8 almost got there, initially). Borland died a sad death, and what we see now is nothing but Post Morten flatulence.

Re:Borland has died after Borland Delphi 7

[Timesprout]

I agree your first point,but if you cant hang on to your top people then serves you right. What really killed Borland though was their pricing structure with their no updates, everything is fixed in the next release which you pay nearly full whack for and IBM releasing Eclipse which was intended to destroy Borlands main revenue source in the emerging J2EE market at the time.

Re:Borland has died after Borland Delphi 7

[pdawson]

They sorta realized this. CodeGear is all the development apps (Delphi, C++, C#, Java, etc.) spun out to a new company. Borland is still around, and they're keeping the code management crap for themselves.

Re:Borland has died after Borland Delphi 7

[ralphdaugherty]

Had kept the focus and the creative minds, either .Net would not exist (and consequently, stole Borland's thunder) or the Borland tools would be better even than the Microsoft ones on that fronts (Delphi 8 almost got there, initially).

      Which is why Microsoft did what they did. They have always done this to their competition, just add it to the long list.

  rd

Re:Borland has died after Borland Delphi 7

[Anonymous Coward]

Excellent posting vivaoporto, & dead-on accurate!

I agree on all counts, and wasn't aware of a couple you made in fact, in Delphi outselling Ms compilers.

Borland Delphi was not only outselling MS compilers (which I have used on my own for freeware/shareware development, but also professionally since 1995 (mostly VB3-VB.NET in Visual Studio 2005 currently though)) but also outrunning & outgunning MS' stuff.

Did you know, Borland had Delphi.NET out before Microsoft did for .NET RAD development? Pretty ironic imo.

Personally, I've used Delphi since 1995 in 16-bit version 1.0, all the way up to 7.0 in 32-bit environs on Win32 (and even Linux via Kylix, the Delphi for Linux), and it kicks butt!

The FASTEST compiler (in terms of compilations speed, unmatchable) & in terms of code it produces, & how fast IT operates.

PROOF?

In (of all places) Visual Basic Programmer's Journal (forerunner of, iirc, today's .NET magazine) October 1997 issue entitled "Inside the VB5 Compiler", Delphi so soundly beat BOTH VB5 and MSVC5&6, it was amazing.

Borland Delphi 2.0 swept the floor with both MSVB5 & MSVC5-6, almost taking EVERY test, & only losing OCX form loads to VB5, & form repaint speed to MSVC++, but the other 8 tests it blew them BOTH out of the water in!

Especially in Math & String processing speeds, which is, as you know, something EVERY program does a lot of.

It is the "WHY" of why I selected to write this program in it (because it heavily uses math & strings!

I wanted an INSTANT performance advantage out of the box is why! Delphi per the results I saw in the VBPJ mag I told you about earlier gave me that. In fact, I turned from using VB & MSVC++ in 1997 because of it, going more to Delphi until around 2003. Ms does get more following and for 'mgt. reasons', not technical excellence. "Microsoft will be here tomorrow, they have the cash: Will Borland?" being the reason.

Funny, but I can write up code in Delphi 2.0 & 3.0 that runs FINE on today's Win32 OS & hardware... idiots @ the wheel controlling things & being bean-counters types only, & not coders, should NOT be leading IS/IT/MIS dept.'s but as you know, it is largely this way, unfortunately.

Anyhow, using Delphi does yield higher performance code than VB & VC++. To go past that, I then I tuned it above compiler options (inline assembler code & multiple thread design usage) & manual profiling later (registering hi-res multimedia timers & ticking off how much time each routine took & worked on the longer ones etc.) homemade, & oldschool, but, I AM, lol, old!):

APK REGISTRY CLEANING ENGINE 2002++ SR-7:

http://www.techpowerup.com/downloads/389/foowhatev ermakesgooglehappy.html

Thanks for these tips, & enjoy the program above should you try it!

(It is the ONLY registry cleaner that works from a SINGLE CODEBASE on Windows 9x/NT/2000/XP/Server 2003/VISTA w/ out modification, and beats any other out there, via tests done on 3 diff. forums during 1998-2002)

"When Borland (then Inprise, then Borland again, then Codegear(?) ) stopped making sober RADs and decided to take a chance on expensive toys for code management, they lost in both fronts. The Turbo Series (Pascal, C and Assembler) and Delphi (the odd versions, 1, 3, 5 and 7) seriously competed against Microsoft products (Microsoft C, Assembler, Visual Series), even outselling them in a lot of places in the world (Brazil, for instance)." - by vivaoporto (1064484) on Wednesday March 28, @08:04AM (#18513581)

I started out on PC programming using Borland Turbo Pascal 5.0, & then lastly Object Pascal 7.0, in collegiate academia settings back circa 1991-1992 for DOS. Good stuff for its day, but a HELL of a lot better in Delphi 1.0 - 3.0 even (5.0 - 7.0 were king/tops as you felt,

Re:Borland has died after Borland Delphi 7

[osgeek]

I'll take it a step further. Borland/Inprise/whatever is such a fucked up piece of shit company that I'd never knowingly start a serious project that depends upon them or their products in any way.

My team has suffered from blistering crotch fires of agony trying to cope with C++ Builder's (5 & 6) linker woes. Rather than spending our valuable development time on important-for-our-customers product development issues, the very existence of our company became reliant upon working around our inability to even link our growing application.

It was a horrible mess, and one that Borland was useless in helping us to resolve because they were off working on new products that never saw the light of day.

Never *ever* again.

Re:Borland has died after Borland Delphi 7

[birder]

Yeah, they died (I'd say after Delphi 6 which I still use today) but Developer Studio 2006 is an impressive product for Borland friendly programmers.

Re:Borland has died after Borland Delphi 7

[Bozdune]

...blistering crotch fires of agony...

Thank you, that's one to file away in the "useful phrases" department. I like it.

Re:Borland has died after Borland Delphi 7

[Anonymous Coward]

Yeah, but you'd think they'd have understood their lesson by now: changing their name to something nobody's heard of (be it inprise or codegear) is a bad idea. Nobody's going to buy it. Mind you, borland (or whatever they like to call themselves nowadays) is long dead in my book. The *ONLY* useful thing they make these days is Together (which is too expensive anyways). BTW, wasn't codegear for sale? Looks like nobody was interested after all...

There's TONS of free IDEs (Eclipse, Netbeans, Sun Studio, the Visual Studio Express Editions, etc), lots of free compilers (GCC, all the .NET ones already included in the framework, etc), SDKs and everything else. Some languages are even totally open source or are going this way (like Java). Borland and their expensive tools aren't free, are as closed as it gets, tend to be buggy (some versions sure were at least), and their "main" languages like Delphi just don't have much of a following anymore (haven't seen a Delphi job in years).

Either ways, there's hardly any money to make with PHP -- the niche of cheap/free stuff (cheap LAMP hosting and all). And it mostly revolves around linux, so a Windows IDE... There's already a bunch of IDEs anyways (lots of free stuff too) -- that don't lock you in with VCL crap unlike this thing, and likely cheaper. And there are LOTS of far batter frameworks than what this uses -- you even get to choose what works for you. This will soon go the way of the Kylix.

Looks like their only real product is going to be that "app lifecycle management" crap, just like IBM's expensive, overhyped and overpriced Rational suite...

For the Amateur coder? NOOB!! ALERT!!!!

[Anonymous Coward]

So I don't know much about history, but I am capable of using Dreamweaver to build web pages that connect to a MySQL database. Is this something worth me looking into?

Re:For the Amateur coder? NOOB!! ALERT!!!!

[kni52]

I don't know about the Delphi PHP IDE in the article, but if you are looking for a new develpoment enviroment, I would strongly recomend trying eclipse http://www.eclipse.org/ with PHPeclipse http://www.phpeclipse.net/tiki-view_articles.php and XAMPP http://www.apachefriends.org/en/xampp.html. This compination is far more powerful than Dreamweaver, free (as in beer and speech), and has plugins available to support almost every every language I can think of. Which will help you in learning new languages with out having to learn a new IDE. Inaddition it can be made to run off a USB drive. Eclipse is java based so it is available for linux windows or osX.

The only advantage Dreamweaver may have is the WYSIWYG editor. There is an eclipse plugin for this (included in easy eclipse mentioned below) but I haven't used this much, since it's geneally faster and more reliable to edit the code to get the result I want. PHPeclipse includes an browser preview pane that, with the exception of directly editing the view in a WYSIWYG maner, is just as functional.

My two favorite features are the integration with XAMPP (an extrememly easy to setup and use local LAMP web server, which should work with Dreamweaver too) which allows for a Dreamweaver style design view for PHP code, and the Remote System Explorer (RSE) http://www.eclipse.org/dsdp/tm/tutorial/ plugin which allows you to connect and work on files on a remote webserver as if they were local, which is great for quick fixes and fixing typos.

To use eclipse, PHPeclipse, and XAMPP to create a portable development environment on a USB drive, check out http://www.plog4u.org/index.php/Using_PHPEclipse:I nstallation:XAMPP_Example_Installation for eclipse and PHP eclipse installation into XAMPP and this thread http://portableapps.com/node/929 to make it all portable. I would also recomend checking out http://www.easyeclipse.org/site/home/ for easy installation of eclipse with many of the most useful plugins preinstalled for you.

Re:For the Amateur coder? NOOB!! ALERT!!!!

[ooh456]

A slight disagreement there. Don't be so quick to slam Dreamweaver. Especially 8.

Eclipse is great if you write classes all day but if you design real world web sites Dreamweaver 8 is at least as good an option as Eclipse.

With Eclipse you get fewer code highlighting options, worse code search, no built in reference, no site synchronization tools, no ftp, no decent visual editor (still useful for those quick pages and/or finding and jumping to the the right code to edit in a template file), and since it's Java it just feels wrong compared to C programs. There are more things lacking as well which are useful for web developers and not just code heads. Dreamwevaer is still the champ for people who can afford it.

The main difference is: Eclipse is aimed primarlily for programmers... Dreamweaver is for web site developers. If you fall in the middle like me... you can use either one but for me it's mostly DW until Adobe manages to eff it up or people develop better plugins to Eclipse.

I'm completely thrown

[Anonymous Coward]

I'm sorry... Delphi for PHP? Am I the only person who read that and thought "ftw"?

Re:I'm completely thrown

[Idaho]

I'm sorry... Delphi for PHP? Am I the only person who read that and thought "ftw"?

No, I was in fact thinking the exact reverse.

Although it's still strange, this thinking in terms of TLA's.

Re:I'm completely thrown

[ibbo]

No I thought that too. I also thought uh, php mysql nice NIX tools yes but a delphi IDE that only runs on Windows.

Bah will never work. If your into webdev on windows your a VS c# organ grinder nothing more.

Ibbo

Oh, great

[Corporate Troll]

Writing insecure web applications in less time. Thanks Borland! ;-)

Re:Oh, great

[Vexorian]

What the heck?

So, your last knowledge of php was about version 3.0 ? And then you just skipped it and you still get the right to make these comments?
The alternatives to php are not any more secure than the current version, sorry about that.

Re:Oh, great

[Corporate Troll]

Didn't you notice the smilie at the end? Go buy some humour...

Re:Oh, great

[Anonymous Coward]

Go suck PHPBB, PHPNuke, Postnuke, PHPwiki etc.

Re:Oh, great

[cortana]

O RLY?

Another desperate attempt.

[Yuioup]

Another desperate attempt by Borland/Codegear to appease their dwindling developer community.

Delphi is dead. Nothing to see here. Move along.

Y

Re:Another desperate attempt.

[dunkelfalke]

no, no, it's resting.

now serious: delphi is still the best tool for developing win32 apps and it is still widely used in europe.

Re:Another desperate attempt.

[Yuioup]

Agreed. Delphi 7 rocks.

Yours truly,

A disgruntled Delphi 8 purchaser.

Y

Re:Another desperate attempt.

[rucs_hack]

Delphi six is the best release of the IDE so far as I'm concerned.

all these years and it's still my preferred instance of delphi.

Re:Another desperate attempt.

[gasmasher]

It may be dying but I still make nearly six figures a year contracting as a Delphi programmer. It's not dead yet.

Re:Another desperate attempt.

[illegalcortex]

Meh. People have been telling us Delphi is dead longer than they've been saying that about Macs. We'll believe it when we see it.

Re:Another desperate attempt.

[davek]

forgetaboutit. Borland made one of the first C compilers, has expanded software development since the beginning, and has no intention of quitting the development game any time soon. Sure, it seems to have its head up its ass with a lot of things (ditching Kylix seems to be one), but the fact remains that there will always be a need for an alternative to MS Visual Studio for IDEs. If Borland were smart, they'd play their advantage: not being locked in to windows.

Delphi is still a solution in search of a problem. I use the thing 8 hours a day and while it does have a tendency to allow poor coding style, that's a shortcoming of IDEs in general and not Delphi. While I wouldn't bet my bottom dollar on its success, I still think that it will become more applicable in the future, not less. Delphi's problems have to do with the idiots who are directing the project, not something inherently wrong with the IDE.

-dave

Re:Another desperate attempt.

[Afecks]

I use Delphi 6 and have done so since it was released. It even works in Vista. I really don't see it being dead as much as it is starting to get a little long in the tooth. I dare not switch to Delphi 7 or later because some projects simply will not build properly and it scolds you for using pointers. Other than that Delphi works a charm. Most components still support Delphi 6 including Indy.

Apparently you have no clue about php ?

[unity100]

With php, code is seperated from the design, templates handling the visuals.

all major software like phpbb et al are like that.

you can make it so that not a single byte of code mingles with template if you wish.

php offers liberty on seperating content with the code. some do not, some do.

Re:Apparently you have no clue about php ?

[FesterDaFelcher]

With php, code can be seperated from the design, templates handling the visuals.

There, fixed it for you. Didn't want your flamy "no clue" comment to get you in trouble.

Re:Apparently you have no clue about php ?

[unity100]

hey - everyone is entitled to go flamy once in a while.

You kidding me. .NET alternative ?

[Shohat]

With all due respect , .NET is a great platform for enterprise-level development. PHP, no matter how you flip it or manage it , isn't . Zend is not big enough to develop and patch a platform that can compete with Microsoft's flagship.
Cheers

Re:You kidding me. .NET alternative ?

[Anonymous Coward]

With all due respect , .NET is a great platform for enterprise-level development.

Hardly. .NET doesn't run worth shit on my company's Solaris and HP-UX systems. Yes, we're talking about real enterprise systems here. Some Intel box running Windows 2003 Server and .NET does not constitute actual enterprise development. Real enterprise work is done on UNIX systems like Solaris, HP-UX, and AiX.

.NET and Java just can't offer the execution speed necessary for really large-scale enterprise apps. That's why we still use C++ in many cases. And yes, I know that in a small number of unrealistic microbenchmarks Java is shown to be faster than the equivalent C++. But in reality, that's just not the case. .NET and Java just don't scale well enough.

Re:You kidding me. .NET alternative ?

[maxume]

There are plenty of web sites running on janky systems that are, ya know, bigger than 'really large-scale enterprise apps'.

Re:You kidding me. .NET alternative ?

[Anonymous Coward]

I hope you aren't arguing that PHP does scale well enough and Java doesn't? That is a ludicrous remark.

Re:You kidding me. .NET alternative ?

[Anonymous Coward]

Read my post again, paco. Do you see any reference to PHP? No, you don't. Everyone knows PHP is inferior to .NET. And what we're saying here is that for real enterprise development, .NET is insufficient. So it should be clear to you that PHP is not suited for enterprise development in any fashion.

Re:You kidding me. .NET alternative ?

[Anonymous Coward]

You can do anything if you throw enough hardware at it. And that's what a lot of those web sites do. Their software implementation is so terrible that they need five to ten times the amount of hardware they should, just to offer a minimal level of performance.

Then again, not every enterprise installation is a web site. There are many financial institutions whose systems process more transactions each minute than some of the most popular web sites handle in a month. Again, .NET doesn't run worth shit on any of IBM's high-end mainframe systems used for such computing.

Re:You kidding me. .NET alternative ?

[umghhh]

but the problems with speed are down to the coding faults people make because they have not a clue how java really works (not a clue about .net - never worked with it). OC you may say that because java folk is an uneducated lot then it is irrelevant whether java code can be fast or not - dead weight of java coder will kill the speed anyway. //

Perhaps not the best acronym for a new project.

[Anonymous Coward]

Oh let me just google up some info on this new VCL thing I heard about and ... oh my god anthropomorphic wolves and foxes having sex!? Nooooooooooo!

Re:Does this IDE build upon the existing interpret

[Aladrin]

It appears to be the regular PHP, but they've created a library for common controls. (Yes, they made a library for making textboxes and radio buttons. -yawn-) At first I thought it was a php plugin written in C or something, but no... It's just plain PHP. What was wrong with making the controls in HTML like always, I dunno.

I'll admit, I've written a function to take an array and make a dropdown box from it, but Delphi's VCL is going a bit far, I think.

Looks Nice!

[purpleraison]

I am a long time Zend user, and I really like ZDE -- especially because it is cross platform (Linux, OSX, and Windows).

Delphi for PHP looks to be very similar (I read the announcement, but have not tested the app yet), but also has a database browser! This is particularly valuable, and truly DOES speed up development.

I will have to blow the dust of my Windows box, and try this out. :)

Re:Looks Nice!

[gbjbaanb]

a database browser! You mean like this one? http://www.mysql.com/products/tools/query-browser/

Re:Does this IDE build upon the existing interpret

[Red Alastor]

It's largely based on the qooxdoo javascript library.

Check the demos here and here.

Did they miss the point?

[muxecoid]

If webdevs do something Windows they use ASP (ASP.NET). ASP.NET is already RAD-like, the niche is taken.
If they offer tools for PHP and MySQL target servers run Linux, target developers run Linux, and they are missing again.

Re:Did they miss the point?

[MedBob]

Have you used .ASP? EWWWWWWWWWW!
It's the typical M$ bailing wire.
This provides (right now) about 75% WYSIWYG development of PHP web pages, using a framework that closely resembles the Object Pascal (Delphi) Visual Control Library (VCL). It gets you a Web GUI prototyped quickly, and an event-driven framework to develop the back end.
It will DEPLOY to either M$ or Linux, so the only constraint on your hosting is PHP 5.

Re:Did they miss the point?

[AutopsyReport]

Actually I think you'd be surprised how much PHP development is done within Windows and later pushed to a Linux server. In fact, where I work (federal government), PHP development and deployment is done exclusively on Windows. So I don't think this product is off-base. Besides, I'm thinking the Windows crowd is more inclined to purchase an IDE than the Linux folks.

Re:Did they miss the point?

[megalomaniacs4u]

Um, no around here we develop on Windows and upload to the linux servers (or even windows servers running php & iis)

There is no requirement to develop php on linux I have no idea where you would get a backassed idea like that...

Downloading the trial

[SarekOfVulcan]

I'd love to give it a try, but it requires a login for download, and there's no "Forgot your password?" option.

(And yes, I have a whole bunch of other email addresses I could use. That's not the point.)

Let's hope you're evaluating REAL REAL fast

[suv4x4]

... because CodeGear offer this trial on their web site:

"Free, fully functional 1 day trial"

Right, 1 day.
Great.

Re:Let's hope you're evaluating REAL REAL fast

[Gramie2]

As one of the CodeGear people explained, it was intended to be part of some kind of 1-day on-line seminar. The seminar had to be cancelled, but they decided to make the demo available. They realize that a single day is "sub-optimal".

Re:Let's hope you're evaluating REAL REAL fast

[suv4x4]

As one of the CodeGear people explained, it was intended to be part of some kind of 1-day on-line seminar. The seminar had to be cancelled, but they decided to make the demo available. They realize that a single day is "sub-optimal".

PR is hard. Basically if the CEO's About page needs a photo but lacks one, which do you think is the better option, PR-wise:

1. Use the only photo of the CEO available, where he has his pants down.

2. Wait a bit and make/provide a better photo.

Unless they plan to assign a CodeGear "guy" explaining the situation to every user visiting their site, they'll look like morons to people who go there to evaluate their software.

Not only is it 1 day, but the activation is ridiculously complicated (activation.. of my trial.. by basically putting files in my documents and settings/[user] folder)

Re:Let's hope you're evaluating REAL REAL fast

[Suidae]

This is why microsoft/vmware provide free virtual machine software.

I use Microsoft VPC with a Win2k image. Install SVN and connect to a repository on the network, set up a startup script to set the time system date to Jan 1st then save and backup the VHD. Adjust the settings file to disable time syncing. When you want to evaluate software, copy (or inherit from) the image. Turn on 'undo disks' and you have the option to discard all changes made during the run.

I call the image 'Groundhog day'.

Re:Let's hope you're evaluating REAL REAL fast

[suv4x4]

This is why microsoft/vmware provide free virtual machine software.

That's like me ranting "BluRay's DRM sucks" and you telling me "this is why hackers provide hacks".

It doesn't make Sony/Toshiba any better for feeding us rogue DRM. Of course I installed it on VMWare, otherwise...

Re:Let's hope you're evaluating REAL REAL fast

[nuzak]

> I call the image 'Groundhog day'.

Shouldn't you set the date to Feb 2 then? Besides, most stuff with expiration these days phones home and won't run if it can't. Pretty hard to fool that.

Re:Let's hope you're evaluating REAL REAL fast

[niff]

Make sure to have "Teach yourself PHP in 24 hours" ready.

Re:Let's hope you're evaluating REAL REAL fast

[Old Wolf]

Not only is it 1 day, but the activation is ridiculously complicated (activation.. of my trial.. by basically putting files in my documents and settings/[user] folder)

If you can't manage to put a file in a folder then I'm glad you aren't going to be developing applications that I might have to use one day!

Re:Let's hope you're evaluating REAL REAL fast

[Anonymous Coward]

If they can't manage to automate the process of putting a fucking file in a folder, then be glad I won't be using their IDE to develop applications that you might use one day!

Re:Let's hope you're evaluating REAL REAL fast

[Suidae]

True, but until they fix it you can have you cake without buying it.

Re:Let's hope you're evaluating REAL REAL fast

[suv4x4]

True, but until they fix it you can have you cake without buying it.

You know, it's not as if I'm dying to have Delphi for PHP. It's more like THEY dying for us to have it, like it, and finally buy it.

It's extremely hard to convince PHP developers, most of whom selected Apache/PHP because it's free, and Apache/PHP hosts cheap, to buy an app to do what they can do better with free IDE-s like Eclipse PDT.

Delphi Dead?

[jshriverWVU]

Ok I'll bite, might have some karma to burn. Isn't Delphi Dead? My first task at the job I have now was to port some Windows Delphi code to linux. At first I thought, no sweat I can use Kylix. In the end I came to the conclusion Delphi is dead on any platform and ended up rewriting it in C. I guess it could still have life for legacy applications, but in this rapid multi-flavored world if a language is 100% stuck on 1 platform/OS it's more or less dead (excluding of course assembly which of course is arch specific).

Re:Delphi Dead?

[Darkstorm]

Hmm, you know, I am making a decent living using delphi. Is delphi dead? Pretty much, but since windows does occupy a vast majority of pc's out there (fact, not fiction), I will have to disagree with your concept that any language that is not multi-platform as dead. Maybe one day when linux becomes more user friendly, and people stop screaming "look at the source" as a replacement for documentation and simple to use interfaces, maybe your statement might be true. There are far too many idiots in the world that will take anything mickysoft hands them and think it's the greatest thing ever (when they are not cussing it for crashing). Overall, I don't foresee being out of work any time soon.

Note: I have started down the dark road of .net, but I still like to play with php, since php and mysql is on almost every low cost hosting company out there.

Re:Delphi Dead?

[jshriverWVU]

That's good. Honestly I liked Delphi. I downloaded the 180day evaluation version of the suites so I could look/compile/test code while porting to C. The IDE is beautiful, I'm sad Kylix didnt make it because it would have been a nice easy way to make X applications without having to know xlib.

Re:Delphi Dead?

[Eli Gottlieb]

Apparently, you've never heard of the Free Pascal Compiler or the accompanying RAD IDE Lazarus.

Re:Delphi Dead?

[jshriverWVU]

Aye I have and they're not 100% Delphi compatible, haven't tried Lazarus though. The problem with using fpc for Delphi code (besides incompatibilities) is that it doesnt support Forms. If you're doing cli only stuff it seems to work ok. But if you have a GUI you're kinda screwed.

Re:Delphi Dead?

[Ronin Developer]

Is Delphi Dead?

First, you have to define what you mean by Delphi. I code using Delphi 5 & 7 (i.e. Delphi's flavor IDE and Object Pascal) every day. With the number of controls available and the knowledge of being able to create my own visual and non-visual controls, this tool allows the company I work for to remain well ahead of our competitors by at least a year ( they copy our features ).

However, Delphi is no longer just the language - it is now a family of IDEs for many different programming languages that have adopted the same advanced IDE and concepts of the original Delphi product.

Is Delphi dead? Well, try to find competent Delphi developers and you'd be suprised. It's probably easier to find older Delphi developers who know the environment very well than to find younger developers versed in it. If you need a Delphi developer, be prepared to pay them well as they are a rare commidity indeed. And, like any developer for any tool/language, their quality and skills vary.

Has Borland/CodeGear blown it? Perhaps. The definitely pissed me off when they raised the price of their tools well out reach of the small developer. They did that when Phillip Kahn built that palace in Scottsdale. Then, they moved to this application lifestyle BS and, essentially, abandoned their core customers. WTF were they thinking? Then, they blew it with Kylix - they didn't fully develop it and keep the costs down to make it easy to adopt. Then, they dropped it like a hot potato. I haven't upgraded my products since then - I certainly wouldn't be able to afford them (the Enterprise and Architect versions) on my own.

Is $249 or $299 too much to pay for Delphi for PHP? Maybe. They will have to show the community that it's worth spending the big bugs over some other IDEs (free and commercial). Will I play with Delphi for PHP? Probably - if they make a trial version that isn't limited (like the Turbo Explorer products are) and actually be able to create my own components and such. And, it sure as hell better be able to talk to Firebird, MySQL and Oracle and not just Interbase - Yes, I like Firebird.

So, is Delphi dead? Let's say that I am retraining myself for C++, C# and probably Java development in the event that I need to change jobs. But, I will continue to use Delphi as it enables me to put food on the table and pay the bills. And, I like it.

RD

Re:Delphi Dead?

[Eli Gottlieb]

Take a look at Lazarus, they're supposedly working towards Delphi compatibility.

Now I admit FPC doesn't have full compatibility yet (even in -Mdelphi mode), but it's still a damn sight better than rewriting in C or C++.

Borland should burn in hell for ruining a perfectly good language and IDE.

Re:Delphi Dead?

[Franciscan]

Borland just released Delphi 2007. It's got first-class support for Windows Vista apps (Glass, etc). How exactly is Delphi dead? There are probably 100,000 developers in north america and probably half a million or more active Delphi developers worldwide. Are we all dead? There are more people doing Java these days, and more people doing C# and probably more people doing C++, but we're not DEAD. Why do you have to be #1 or be called dead? Warren

Re:Delphi Dead?

[Directrix1]

Few code straight to xlib. They usually use a toolkit like GTK or Qt.

Re:Delphi Dead?

[Darkstorm]

I fiddled with Kylix, and I've played with most of their development tools. It's funny how many people stopped at delphi 7 and ignored the later version. Borland lost sight of what kept them in business, good usable tools, and that started somewhere around delphi 8.

I have used zend's ide to work with php...can't say it is the best ide I've worked with, but it did the job. I still have to install the one day trial (probably on the weekend so I have a few extra hours to play with it), but if it is half as nice as delphi 7, it could be worth the money. Till then, I'll reserve judgment.

Re:Delphi Dead?

[Halo1]

The Free Pascal Compiler is indeed only about the non-GUI stuff. All the GUI stuff (including the forms unit) is handled by the Lazarus team.

Re:Delphi Dead?

[suv4x4]

You know you're pretty good at asking yourself tons of questions and answering right away. You're probably never lonely.

For the record, a dead language is never THAT dead that everyone suddenly drops it. Fortran, Lisp -> they are still used in lots of projects out there but they're pretty much all dead.

In fact, you can argue PHP is starting to die this year, but it still has a majority as a server side technology on the web - it's simply not as simple as saying "is it dead? I use it, but I'm ready to jump ship" and mean something concrete at all.

Re:Delphi Dead?

[Ronin Developer]

Listen - I am a big fan of Delphi and I've been onboard with it even before 1.0 was releasesd. I'll fight for its use over just about anything else out there. However, do a DICE or MONSTER search on DELPHI and tell me what do you see? On DICE.COM alone, nationwide, there are 183 jobs posted in the past 30 days. Compare that with JAVA (16K+) or C# (8K). But, there are 183 jobs - so, somebody's using it.

Of course, it could simply be that we Delphi developers like our jobs sooooooooo much that we don't leave our employer and thus, they have no need to hire more developers to replace us because our productivity is so high. (yeah..that's the ticket). Or, it could simply mean that Delphi is not a hot commodity in the US. I also looked at the salary and consulting rates ($45/hr) - which translates to roughly $90K (before taxes). We don't come cheap, do we? What does a comparable entry level VB, C# or C++ developer command?

World-wide, I agree, there is a large user base for Delphi and Borland products. I prefer Borland products myself. I've used Delphi, JBuilder, C#Builder and C++Builder. Lest we forget, Java Beans came about because of Borland's involvement. I believe the original Delphi, C#Builder and C++Builder now fall under the title of "Delphi". I feel Borland IDEs are superior - but, then again, look at what happened to BETAMAX? I liked BETAMAX - it was clearly superior over VHS. Yet, which format won out? Whose IDEs are more popular for the Win32/64 environment, Borland or Microsoft? Why?

Are we dead? Is Delphi dead? Hardly. But, Borland / CodeGear has do something to improve their image. They need to get their heads out of their proverbial butts and come up with a real marketing plan. They need to find a way to win back the core developers they alienated over these past several years. They need to find a way to get the attention of developers like the average reader on /. and convince them to look at their products again if not for the first time. They have a lot of bias to overcome - just review all the negative comments about Borland that have been posted here over the years. Only if they can do that, Delphi, as a mainstream product, can be resurrected.

RD

Re:Delphi Dead?

[Franciscan]

Yeah. You're right. 100%. However, comparing Delphi marketshare to Java marketshare is kind of pointless, because Java has something Delphi will probably never have: portability. Remember Kylix? Didn't fly. There are both technical and market reasons why. You can't just compare Windows-dev-tools with portable dev tools, when comparing market share. It's like saying "the Toyota Corolla is selling well, therefore there is no market for the Porsche 911". Java and Delphi are only competitors in a very small way. How many Java developers develop commercial Windows apps in Java? Not very many. Java is an in-house-software-systems-language, and web-services oriented system. All the stuff people write in Java is stuff I'd never write if you paid me 10x what I get paid right now. BORING CRAP. Sorry, but there it is. Not all developers are the same, and not all development projects are the same. Delphi's core constituency is formed by people who write software for Windows only, and who want software that runs on every version of windows ever made (even back to windows 95, or windows 98, if you really need to) with a single binary EXE file, with no runtimes needed, no necessity of using DLLs, and so on. Delphi has two major streams within that core-constituency, in-house-app developers (like myself) that write vertical line of business applications that a company either uses, or sells to its small customer base. My apps have to run on customer PCs around the globe, in every country in the world, without DLL hell, without runtime (sucks to be .NET) and on every version and every service pack of Windows ever released. Even Microsoft can't do this with their own tool set. Oh, and they abandon their own developers (hello VB 6 fans, how ya doin?). So your choices are: 1. Abandon your Windows-only strategy, go Java or something portable. (not a bad idea.) 2. Use whatever latest pile of CRAP Microsoft is shipping these days, and be prepared to rewrite when they ship you a whole new steaming pile of crap. 3. Have an insulating layer between you and Microsoft's platform. That is what Delphi/VCL gets you. If you need to write software that runs on Windows, CodeGear is the only thing I would use. Otherwise, get me outta here. The day I can no longer work as a Delphi developer, I get a new career, or maybe I switch to Cocoa/ObjectiveC on Mac OS X. At least it's elegant and beautiful, and not a stinking pile of MS-crap. I'm bitter today, don't I sound bitter? Yeah. But there is a lot of hope. CodeGear is spinning up now, and they've shipped my favourite Delphi version ever (already), Delphi 2007, and Delphi for PHP, both of which are amazing products. Now, let's see if they can market it. That's where the old BORLAND always sucked. Great on technology delivery. Crap marketing. Franciscan

All well and good till they kill the product...

[codepunk]

All well and good, till they kill the product like they did with Kylix...then what, yep you are left holding the bag....thanks, but no thanks..

I want it yesterday!

[billcopc]

Funny how just yesterday I was digging through my binders in search of my Delphi discs to code up a quick file management util. I'm eagerly waiting for the trial version of Delphi PHP to finish downloading, but if it's anything like Delphi/BCB I'm going to need more kleenex. Say what you will, but when it comes to rapid prototyping it's about as fast as it gets. There's a certain elegance to ObjectPascal despite its simplicity, and it can do just about anything C++ can, with less headaches. I see it as a middle ground between VB and templatized C++, plus the compiler's crazy fast, perfect for the kind of incremental debugging that comes naturally with prototyping.

If they've managed to accomplish this with PHP/Ajax, we'll be seeing a new breed of web apps, coded not by Java dweebs or Ajax hacks, but by seasoned app designers... a little less flair and CSS shenanigans, and more direct interactivity and usability.

Re:Does this IDE build upon the existing interpret

[Shulai]

HTML was wrong because if you write the stuff yourself they had nothing to sell.
Worst part, while I'm not sure if you can ever stop using Delphi for PHP and still be able to further develop your app, or become trapped in both the IDE and the Windows platform.

Re:Does this IDE build upon the existing interpret

[Aladrin]

A good IDE is worth its weight in gold. It doesn't need helper libraries to make it great.

I think you are right about being trapped in it, though... The VCL, if nothing else, would trap you irrevocably. But then, they're treating it more like an IDE for different language, and being tied to a language is true for any language. Just like Ruby on Rails is treated differently than plain Ruby.

Your app on Delphi

[suv4x4]

This is your app on Delphi (TM):

I installed the trial Delphi for PHP and created an app which prints "Hello World" on the screen.

For a reference, this is how this looks in plain PHP (granted no MVC and so on, but for the sake of example..):

&lt?php echo "Hello World" ?>

What does Delphi do?

1. Loads several thousand lines VCL code
   
2. Loads all the menu, form, container and "external" controls, although they're not used (thousands of lines of code)
   
3. The Hello World is a label (no simpler way) which has around 50 properties (color, bg color and what not) defined in an XML file. I left all at defaults, but never mind. The file is loaded, parsed.
   
4. The Label class inherits from CustomLabel, which inherits from Components which inherits from other stuff I didn't even bother check, it goes through all properties, and figures out after a lot of thinking that it should print the words "Hello World".
   

Keep in mind I simplified it so you stay with me. There's also a bunch of other stuff happening, application classes and what not.

And again, this is how it's done in plain PHP:

&lt?php echo "Hello World" ?>

This Delphi stuff is really promising I tell you. Or, rather, it's supposed to look promising when Borland pitches CodeGear for sale again. Don't forget, CodeGear was spun off so that it's income is more clearly defined, and it's a more lucrative sale. Borland doesn't care of CodeGear has a future, it only wants to make it LOOK as if it has a future, and this project is sadly nothing more than this.

Re:Your app on Delphi

[ajs318]

Well, this is what happens when you use an authoring package of any kind. It's a pretty much unavoidable consequence of using a tool to do a job for which it was not designed. That's the difference between a calligrapher and an idiot with a stencil.

It's obvious when a web page was made using Dreamweaver, because you'll get things like <font> tags around spaces and sometimes nested <font> tags rather than declaring the colour, size and typeface in one {or, preferably, doing it properly with CSS}. The problem is that these authoring packages don't reduce things to their simplest terms the way a good programmer will do as a matter of course. It's rather like using a slide rule to do a series of multiplications followed by a series of divisions, as opposed to alternating multiplications with divisions {which is almost twice as fast; a multiply-and-divide operation is just as quick as a single operation, unless it goes of the end and requires repositioning of the slide}.

All in all, a WYSIWYG front end to a non-WYSIWIG process always ends up being a bit like having a device with piano-like keys that clips onto the neck of a guitar and frets and strums the strings according to the keys you're pressing, so that -- in theory -- a pianist can get a tune out of a guitar. In practice it looks riduculous and sounds mediocre at best.

Re:Your app on Delphi

[Shados]

You're right, but last I checked, newer versions of Dreamweaver used CSS by default (heck, even Microsoft Expression does) and automatically remove or merge pointless nested tags, on top of having a bunch of analysis tools to clean up code. If someone knows how to use Dreamweaver, you'll be hard pressed to tell that they used it, except maybe for the dead giveaways in the auto-generated javascript snippets...

I didn't use Dreamweaver in a long time, but heck, Expression (and Im fairly sure Visual Studio Orcas too) can put all styles in a separate CSS file like a normal programmer does.

Its not perfect obviously, but these tools came a long way since Frontp...sorry, it hurts too much, I can't even type THAT name...

Re:Your app on Delphi

[suv4x4]

Well, this is what happens when you use an authoring package of any kind. It's a pretty much unavoidable consequence of using a tool to do a job for which it was not designed. That's the difference between a calligrapher and an idiot with a stencil.

Actually this is the difference between embracing the capabilities of the platform you're targeting, and blindly porting abstract concepts from one to another and expect good performance.

Guess which of those two examples Delphi for PHP is.

I mentioned in another post I also develop and support a similar framework (for personal project, I don't plan to /yet/ open source it, but who knows).

Did I just accept that using "an authoring package" should come with horrid performance and bloat? No. How do my components work? The component classes are *design-time* only, at runtime, they generate very light, and very specific procedural code to emulate the desired behavior at runtime.

What does this mean in practice. It means that my Label component may have 500 properties, but my framework won't render a 500 line XML and parse it through convoluted component classes at *runtime*. Instead the first time the page is loaded, the component class loads, reads the properties and compiles the "form" (called "view" as in part of MVC in my case). The final code from my framework in the above example?

<php echo "Hello World" ?>

Magic? No, just embracing the platform's strengths and weaknesses.

It's obvious when a web page was made using Dreamweaver, because you'll get things like [font] tags around spaces and sometimes nested tags rather than declaring the colour, size and typeface in one

Dreamweaver in fact doesn't do that for a few versions now. It has completely embraced CSS. It will aggregate your styles in a [style] block in [head], and with a single click (literally), Dreamweaver CS3 will move it to an external CSS file.

Re:Your app on Delphi

[niff]

You're totally missing the point about RAD.

If you want performance and hardly any overhead, then go down a few layers of abstraction and code in assembly.

Even your example needs a lot of bloat to show a simple text on the screen:
An Operating system, a web server, a browser, a network, an entire PHP module, etc.
That's a lot of overhead just to get a simple text on your screen.

The point is that you, as a developer can do things faster using RAD. The underlying layers will take over part of your work as a developer, resulting in faster development time (improving maintainability and scalability), sometimes at the cost of performance.

In the real world development time is often more expensive than hardware upgrades to compensate for non-optimized code.

Re:Your app on Delphi

[suv4x4]

You're totally missing the point about RAD.
If you want performance and hardly any overhead, then go down a few layers of abstraction and code in assembly.

The thing is, I don't have to code in assembly, I simply have to not code in Delphi for PHP and opt for light PHP framework OR .NET OR Python OR Java etc etc..

You'll do yourself to acknowledge there's more out there than "assembly" and "Delphi for PHP" when discussing those things. The "code in assembly" argument is tired, cliche and plain wrong.

Let me tell you why Delphi RAD works on the Desktop:

1. You have the CPU of entire machine to run what you want. You absolutely need 200MB of RAM for that blinking LED effect on top? Big deal, take it, it's there. The important thing is it works, and the entire machine is at your disposal.

2. Delphi for Desktop compiles to Win32+machine code (hella fast) or .NET (also sufficiently fast), both platforms which are over 100x faster than comparable PHP code (yes 100x is not overestimation, chekc the language shootout benchmarks). This means there's lots of headway for Delphi to implement time-saving abstractions and components, as the code runs fast enough to make up for it.

Let me tell you why Delphi for PHP doesn't work:

1. Unlike the Desktop, you're operating under heavily reduced resources. Everyone that uses your app calls back the same server, the same server serves the app to many users. Compare this to a Windows box with say, 100-200 mice and 100-200 instances of the RAD tool opened, and being used simultaneously. Does it still look as if performance doesn't matter to you?

2. PHP is slow as hell. This is not Win32, it's not .NET. Put whole VCL-worth of abstraction on top of PHP, and it becomes slower than hell. Now run it on a busy web server. Does it look promising yet?

If you can't see my point, then I give up. But I'll have to tell you that as a PHP developer, a big chunk of my time is spent optimizing performance of PHP applications, and let me tell you I'd not touch Delphi for PHP with a 20 foot pole.

In the real world development time is often more expensive than hardware upgrades to compensate for non-optimized code.

I did some benchmarks. Delphi for PHP apps will be able to handle around 10-20 people at once on a moderately powerful server. Here's to hoping that your RAD apps don't ever become popular.

And let me put things in perspective for you. Upgrading a hardware is easy. Managing 20 times more servers due to poor web app performance isn't: 20 times more support staff isn't cheap, 20 times more administrators aren't cheap, 20 times more backup data isn't cheap.

Development time IS expensive, but when the result is mediocre, you've just wasted all this development time, and a ton of other resources in your company that could go for a better purpose.

There are RAD frameworks for PHP out there much more robust and performant, which will save you just as much development time as Delphi for PHP possibly could.

"Hey it's RAD so if it sucks, it doesn't suck" is a poor argument in defense for Delphi for PHP.

Don't see this working for PHP

[El_Muerte_TDS]

I like Delphi a lot. It takes a lot of the GUI developing stuff out of your hands, etc. But I don't see this same development method working for PHP.
Simply because they use the same "application" aproach that Delphi had. But PHP, or at least webapplications, are not really persistent. Every time you need to save and restore your application session, and for good performance you want to keep this as minimal as possible. When everytime the "program" has to do something you need to restore the "application" it put some heavy load on the webserver/app in total. Also, unlike popular believe, PHP is not about presentation, while Delphi is quite a lot about presentation. Positioning elements on your form/webpage is something you do through HTML/CSS.
Delphi for PHP is sort of a competitor to Dreamweaver. And on the programming side I think Delphi for PHP is much better, but the design part if probably worse. Also, no serious PHP programmer uses Dreamweaver to develop PHP applications. I don't think Delphi for PHP will ever be a way to create RIAs. I think they would have a better chance when they made a Delphi for Flash/* (where * is a server side thingy like ASP/PHP/Perl/...). At least Flash features a more Delphi like environment, and Flash could seriously use a better interface for "application" development.

One day trial?

[walterbyrd]

Is this a misprint? Are they kidding?

http://www.codegear.com/Downloads/TrialandFreeVers ions/Delphi/DelphiforPHP/tabid/250/Default.aspx

Delphi Developer/Programmer checking in

[dmcooper]

I think this is pretty interesting. I'm running Delphi 7 Enterprise at my office... downgraded from 2005 because it sucked all kinds of ass. I recommend Delphi to anyone who wants an easy RAD that is powerful as well. Delphibasics.co.uk is a good place to get your language basics out of the way.

Re:Does this IDE build upon the existing interpret

[Franciscan]

That's not true. The IDE Product is commercial, the VCL for PHP library is open source (to be hosted on Sourceforge, should be there any day now). YOu aren't trapped. You can go back to VIM at any time. Warren

PHPeclipse

[TwilightXaos]

You probably already know this then, but eclipse can be used to develop php as well:

PHPeclipse User Manual

Re:PHPeclipse

[Abasher]

Or (alternatively) using the official plug-in (which is supposedly better): http://www.eclipse.org/pdt/

mod me 'luddite'

[steveoc]

Please mod me down as an old fashioned technology-phobic luddite .....

But I read TFA, and viewed the demo vid, and I cringed.

Maybe Im getting old, but Im perfectly content writing my PHP code in vim, and trusting that my template/rendering classes that I rely on will automatically look after the 'drawing of the screens' part of the application, in an efficient manner.

Im happy just writing code that twiddles attributes, performs calculations, and calls SQL. The only 'visualisation' that happens during coding time happens in my head. If you need to pull in the description of an SQL table at coding time - just :!! out to a shellscript that generates a template given a table name. Its not rocket science.

The mental state of mind that you need to be immersed in whilst coding is very different to the state you need to be in when testing, or viewing the result from an aesthetic POV. Coding belongs in a text editor, and anything else is a distraction.

Even Ajax - Im perfectly content coding that longhand. Its only a few pitiful javascript functions after all, and I dont see the need to wrap them in a framework. Lets not go around pretending that because we are using AJAX, that we are super-coders on the cutting edge of technology .. its just a few javascript functions and a bit of PHP on the receiving end. You should be able to code that in your lunch break.

OK, so my vim/PHP environment might put me back in the dark prehistoric stone ages, but at least I can sleep well at night knowing that none of my webby code is dependent on the fate of a 3rd party commercial product. After all - thats the main reason I use FOSS in the first place. The whole world wide economy can collapse in a radioactive heap tomorrow, and it wont affect my development at all.

And surely to goodness, isnt vanilla PHP with the standard libraries already way high level enough ? What sort of sheer sloth and laziness leads one to think that they need to front-end PHP with something even higher level ? Are we evolving into a race of Jabba-the-hut's, or what ?

Anyone that commits the blasphemy of 'developing an application' using mostly mouse-clicks honestly needs to be placed into a jar of isopropyl alcohol, and donated to medical science - it is just plain wrong, and always has been.

I tend to take the machine's side of the argument anyway - the less code the machine has to munch through in order to come up with any given result, the happier I am. The end result is just pixels on a screen when you think about it, and a lot of frameworks just add more and more layers of code munching for the machine to produce those same pixels and same behaviour. Silly - just keep it light, simple, scalable and avoid dependencies on proprietary products.

Whats so hard about that ?

Re:mod me 'luddite'

[davek]

Here Here! Great post.

Anyone who relies on an IDE to do programming, in fact, not a programmer. Maybe, someday in the distant future, visual tools will have evolved to the point of being all-powerful, but I don't think that will ever happen. It goes back to the same basic argument (and the fundamental flaw in all microsoft designs, but that's another story): if you want to communicate something, you talk!

VIM (or *cough* emacs) is to programming as the command line is to data processing. Without these tools, we are relegated to sewing our mouths shut, pointing at things on the screen and going "mmmm! mmmm! mmmmmmmm!!!" The only way to talk to a computer is with language. Text editors and command lines aren't the old way, they're the only way.

If I want my friend to pass me the salt, I say "pass me the salt please." I don't point at it and click.

-dave

Re:mod me 'luddite'

[ajs318]

You're talking dangerously like an old-skool hacker there. Next thing, you'll be shaving off a few CPU cycles by omitting extraneous whitespace in text files. I bet you do a bit of electronics or plumbing or something constructive with your hands when you're not using a computer.

Things like this aren't meant to appeal to our sort (I bet you knew what the "logs of sines" and similar obscure tables were used for), but to those pointy-haired types who believe management is a transferrable skill, essentially abstract of the job being done by the workers, and who choose to believe that a single complex tool which claims to be able to do many jobs is better than a selection of simple tools which do one job unconditionally. Which do you think will last longer and perform better: a "magic hammer" which has a camera for identifying different screw recesses -- slotted, Torx, Phillips, Posidriv, hex, tamperproof and so on, a mechanism which automatically deploys the correct driver bit into the chuck, and another mechanism which transforms the impulsive energy of a hammer-like blow into rotary motion; or a set of screwdrivers?

Meanwhile, we can only watch the rise of the New Consuming Class with ill-disguised bemusement; and derive some comfort from the fact that when it all comes crashing down in a heap, we who do actually know how to do something vaguely useful will have the best chances of survival.

Re:mod me 'luddite'

[heffrey]

It's not that it's hard the way you do it, it's just slower.

IDE's don't allow you to write software using "mostly mouse clicks". If you think so then you haven't used a decent IDE.

Re:mod me 'luddite'

[wfberg]

IDE's don't allow you to write software using "mostly mouse clicks". If you think so then you haven't used a decent IDE

In fact, most of the functionality of an IDE should work pretty much independent of whatever libraries and clicky widgets you'd be using. To whit, taking care of CVS, taking care of builds, navigating packages/modules, setting breakpoints and doing the step-through thing while debugging, context-highlighting (oh noes, vim does that, that's evil), predictive input/listboxes that let you pick methods/variables that actually exist/are in scope preventing you from making typos all the time, refactoring, pretty-printing, etc. But for the last 3 features, these things have existed since ye olde times of Turbo Pascal 4 or thereabouts.

You can have a project that compiles just fine using only stock tools from text source files, but use an IDE to make developing a bit (a lot) nicer. Vim and emacs are nice, sure, but they're general purpose text editors. Kinda of like painting the mona lisa with house paint and 2 inch brushes. It can be done, but why would you bother, given that there are specialized tools?

Cry havoc and let slip the dogs of snark

[EmbeddedHack]

Interesting to read all the bwahaha's in reply to the article. It's true that Borland completely screwed the pooch with it's business strategy. It's also true that Borland had a better approach to tools than Microsoft ever did, at least until the 'bidness' types took over.

Laugh all you want - but in the Microsoft world your choices are .net (interpreted bloat), MFC (useful like a box of hammers), or straight WIN32 (is a box of hammers). You've always been at the mercy of the MS market positioning with MS tools. But hey, at least it's a somewhat effective lifetime employment strategy for folks who like to program with hammers.

helping some ppl out.

[Anonymous Coward]

It looks like a LOT of the people posting here could use some help.

Delphi = an IDE
Delphi = also a language derived from Object Pascal.
Kylix = delphi for linux (yes the IDE and the language)
Delphi for PHP = a RAD tool/IDE for PHP

So to all the ones saying "OMG LOLZ DELPHI IS DEAD MOVE ALONG" well, this is an IDE.
To all the ones saying "they'll kill it just like they killed kylix" well, this is an IDE
To all the ones saying "insecure/shitty code faster" --> thanks for sharing your shallow analysis, please continue watching american idol

thanks.
this is an IDE ppl, an IDE. It does happen to have the tedious VCL concept, but geez, it's just a tool, don't worry you won't have to learn scary BEGIN/END keywords to replace { and }, it will be fine, I promise it won't hurt.

Re:helping some ppl out.

[Anonymous Coward]

Yep, anyone who has ever used any other language besides PHP must be a stupid, american idol watching, mouth breathing moron. I'm sure the fact that some shit for brains on slashdot who couldn't code his way out of a paper bag thinks that PHP is the r0xx0rz will really put those programmers in their place. Who are they to judge a programming language as bad just because it is demonstratably worse than every other language that has ever existed in the history of computing?

Borland Missed the Boat

[Tablizer]

I think Borland should have written a GUI-over-HTTP browser using Delphi. Instead of an HTML browser, it would use a GUI-oriented protocol/language that had real GUI features instead of the clunky HTML+DOM. The server-side code would then control the GUI using some kind of XML language. That way it wouldn't matter what server-like language you used. (It would target mostly intranets, not public internet apps.)

Borland wouldn't make money off the browser itself, but rather those who buy Delphi to customize the GUI-browser for their own needs. It is sort like a modifiable Flash-forms engine.

The intranet market is clamoring for real GUI's that don't need software explicitly installed on the desktop (at least not for each app).

Re:Borland Missed the Boat

[Anonymous Coward]

Better yet, If they allready did a delphi compiler backend for .net bytecode....

Why they dont just add a new backend for flash player bytecode? (and flash remoting support to their web class library)

You fucking IDIOTs

[Shohat]

Who the hell modded me Flamebait ?
This is an opinion of a person that actually wrote in PHP and .NET, and I live in Israel and am actually familiar with Zend.
Yes, Java is an alternative to .NET . No , PHP isn't.
Only on slashdot you can find Anti-MS kids running WinXP modding comments about technologies they dont understand.

Re:You fucking IDIOTs

[Anonymous Coward]

Okay, that's great. We don't care. You lost all of our respect the moment you wrote ".NET is a great platform for enterprise-level development."

It's clear from that statement you have never done any serious enterprise development. For if you actually have done some, you never would have made that comment.

Delphi Mortis

[illegalcortex]

This thread brings up the usual "Delphi is dead" or "well, that about does it for Delphi" type comments. As I posted before, Delphi developers have been hearing that for almost a decade. The very fact that it's not dead should tell you something.

The thing about Delphi is that it has ALWAYS been a niche product. Ever since MS started really pushing their coding tools and the other code tool creators started dropping, Delphi has been stuck in a box. But the very fact that it HAS still survived should show you that there's a reason. It is a damn good language and solves business problems in ways that can be FAR superior to anything MS has put out. It's just generally a less painful process.

However, pretty much everything else that Borland has done has failed because there is no nice for it. C++ Builder had some success, but the people going that route were already firmly entrenched in MS or open source products. The same is true for their Java IDE. Then they tried getting into linux with Kylix and Qt stuff. Again, no market. Their C# and .NET stuff has been about as unsuccessful, again because the people doing those things already have a way they want to do it. And when these products don't have a big market, they get neglected. And when they are neglected, even those who would have tried to use them are turned off by the crappy quality. So they wither and die, while the read Delphi remains. Even with all this waste effort dragging the company down, Delphi remains.

Sadly, it appears CodeGear is going to follow in their footsteps. I give the Delphi for PHP thing a year before they stop updating it. Then another year or two before they drop it altogether. Yet again, it's a product without a market.

Re:Does this IDE build upon the existing interpret

[ravenlock]

Thing is, I don't see why I'd want to leave Vim in the first place.

Requires PHP 5

[Danborg]

I downloaded the ONE DAY trial (Ahem!), but couldn't get any sample apps to deploy because my web host is only running PHP version 4.4.4.

Re:Does this IDE build upon the existing interpret

[MrNaz]

How much does an IDE weigh and where can I exchange it for some bullion?

Re:Does this IDE build upon the existing interpret

[FuzzyDaddy]

A good IDE is worth its weight in gold.

Agreed... of course, software doesn't really weigh anything.

Neat but I want it better

[Qbertino]

First of all: All Delphi bashers please shut up. We know Delphi and the IDE concepts it represents are 15 years old but nevertheless it is a neat tool. Some of the best developers I know still like to use it. And if you don't remember the time when Borland jBuilder and it's other IDEs where like 10 years ahead of the rest then you're not qualified to rant. End of story. Thank you.

On Delphi for PHP:
I watched the Screencasts (by some indish guy with a heavy accent). The features are cool and developing stuff goes really snappy. I liked the Ajax components shown.
Then again it has downsides:
a) Windows only.
b) No integrated debugger - the other notable Windows-only PHP Tool 'NuSpere PHPEd' has a fully integrated webserver with a preinstalled runtime and callstack debugger + profiler. It's tough to beat that.
What I kind of liked:
I like the integrated Interbase/Firebird stuff - out-of the box views and stuff inside an integrated DB admin UI. Sadly without a grafical ERD interface. I also like the visual form layout tool. Allthough my designer would probably kill me if I'd use it in a project. :-)

For me it boils down to this: While I - amongst other PLs - do PHP for a living and thus have no trouble spending 300$ or more on a productivity tool that does it's job better and faster than EasyEclipse, the stuff they offer just isn't good enough.

I want Zope with intergrated full-blow Ajax (YUI or something) and a web based developement interface with quasi-grafic ERD, View and Object assembly. All in PHP so it runs on the most shoddy webspace available. With the looks of the Joomla Backend, so my eyes don't bleed after 5 minutes. With a Java of Flash client if needed.

The sad thing I have to tell Borland and Nusphere is that all this will probably be finished by the OSS community even before they port their tools to Linux and OS X. We've got Symfony, CakePHP, Prado, Yahoos YUI, and ten bazillion other PHP and Ajax kits, not to mention the hysteric Rails crowd. Plus the Zope and Django people. We've got KDevelop and Eclipse and Netbeans and Anjuta and Erics Python IDE and Whatnot competing over who's got the larges dick and when MySQL AB finally get's their shit together we'll have MySQL Workbench for free which will give MySQL 5 yet another big push forward and scare the living wee-wee out of Oracle and Co. .

The world simply is moving to fast for me to just buy an IDE that forces me back on to the plattform I hate on a hunch. Mix Delphi for PHP and NuSpere PHPEd and add a visual case tool. Then I'll consider spending 300$ and installing Win2k again. Until then I'll stick with EasyEclipse for PHP on OS X. Even though it's a slowpoke at times.

My 2 cents.

Re:Does this IDE build upon the existing interpret

[Aladrin]

Shhhhhh... It sounds better if they don't think about that ;)

Compiler scoldings

[Latent Heat]

You just need to Project Options, Compiler Messages tab and uncheck Platform Symbols along with the group of Unsafe blah-blah warnings at the end of the list. This gets rid of the stupid scoldings about Windows API specific (that is the whole fine point of Delphi code) warnings.

If you are using Delphi to build ActiveX controls out of VCL controls, you get these stupid [Warning] TF32Ax_TLB.pas(2390): Method 'InitiateAction' hides virtual method of base type 'TControl' messages that I never figures off how to suppress. You really should be able to get clean compiles, either by fixing the problem or using the appropriate suppression flag if you want to accept the warned condition because compile cluttered with warnings you ignore can have you overlook something that really is a warning.

By the way, if you are still using Delphi 6, there was some kind of service pack to it a while ago that fixed a memory leak with Variant arrays or some such thing -- one really shouldn't use an out-of-the-box Delphi 6 unless it has been patched for this.

Re:Compiler scoldings

[Afecks]

I have the free personal version. Any ideas where to get patches for that? Borland's site is a maze, especially now that they've branched into CodeGear.

Re:Compiler scoldings

[Latent Heat]

You can start with http://info.borland.com/devsupport/delphi/, and I tried the Delphi 6 Update Fixes link, and I attempted login for registered user of Delphi 6 (they e-mailed me my password based on my e-mail), and I got a broken link. Maybe that is why there is so much Slashdot hostility to Borland/Codegear/whatever.

The files are D6_Upd2_Pro.exe, D6_RTL2_Pro.exe if you have the Professional version, some other name than Pro if you have the Personal Edition. I have seen sites referencing both the Pro and Personal edition Upd2 and RTL2 patches, but all of the links to them are broken, and I am sorry I am not of more help.

Parent not "insightful" MOD DOWN

[LardBrattish]

I program PHP on a Windows client (In Dreamweaver) & send it up to Linux servers. As a long-time Delphi programmer this is absolutely what I want installed on my Windows PC.

Re:Don't see this working for the web {flame}

[giuntag]

Absolutely agree with parent.

Lots of people nowadays will be happy to tell you how there are 100 (free) MVC php frameworks and templating systems out there that allow "proper" enterprise web app coding instead of the ugly db-logic-presentation mixup that php developers enjoyed yesteryear.

The fact is that much of that framework craze is pure hype. And the java web development crowd should testify to that: how many frameworks du-jour have seen the ligth to be abandoned a couple of months / years later? JSF, javabeans, portlets, you name it.
All the while quick-n-dirty php conquered the web.

IMHO the main error is trying to abstract too much out of the HTTP(+HTML+CSS) model: OOP guys love to pretend they're coding inside a GUI app, and let the magical framework do the translation for them. Then one day they wake up and the realize that either: for drawing one html page that framework issues 1000 ajax calls - bad for scalability, customer gets real pissed, project canned, or: that shiny new css property that would allow page design to be 100% perfect is not part of the framework, sorry dude, or: no way to update only part of your web page via ajax without completely breaking the framework - it's his job to compose the page, after all, or: the html the framework spits is not firefox compliant, etc...

Keep It Simple folks!

php is dead easy for the web because it follows spot-on the web model. One html page = one php script. GET, POST, COOKIES always accessible within your app. Stick as little data into the session as you can. Let the VM clean up at the end of every single request all junk the programmer allocated. etc...

AFAIK, delphi for delphi had this kind of design-with-a-gui-we-translate-to-html power years ago, and I've never seen a single web app deployed like that. .Net takes the same path, and has not been booming, except inside all-ms shops.

Plus, people that do not know coding use flash already. Little value added in switching.

Wow, talk about ignorant.

[Anonymous Coward]

Your example is presentation. Its not mixing anything. Doing it the way you show happens to be longer and harder to read than doing it in a sane way, but its not violating any seperation of concerns. You shouldn't be worrying about what some CS grad would think, most of them are just as ignorant as you. Instead you should be worrying about trying to learn as much as you can to improve your own ability to write secure, reliable and maintainable code.

php

[Ozgur Uksal]

I'll take it a step further. Borland/Inprise/whatever is such a bad company that I'd never knowingly start a serious
project that depends upon them or their products in any way.

Never *ever* again.

When Borland (then Inprise, then Borland again, then Codegear(?) ) stopped making sober RADs and decided to take a chance
on expensive toys for code management, they lost in both fronts. The Turbo Series (Pascal, C and Assembler) and Delphi
(the odd versions, 1, 3, 5 and 7) seriously competed against Microsoft products (Microsoft C, Assembler, Visual
Series), even outselling them in a lot of places in the world (Brazil, for instance).

Two things made Borland wreck their scene: 1) losing their creative minds to Microsoft, specially Anders Hejlsberg
[wikipedia.org], creator of nothing less than Turbo Pascal, Delphi and main architect of C#. 2) losing their focus (from
useful RADs to expensive but totally good for nothing "Application Lifecycle Management" (whatever it is).

Had kept the focus and the creative minds, either .Net would not exist (and consequently, stole Borland's thunder) or the
Borland tools would be better even than the Microsoft ones on that fronts (Delphi 8 almost got there, initially).
Borland died a sad death, and what we see now is nothing but Post Morten flatulence.

ozgur uksal
http://www.adobe.com/

BUY BORLAND/CODEGEAR PRODUCTS NOW !!!!!!!!!!

[Anonymous Coward]

Codegear are not that bad ! They have changed their name !
That is a step forward. True, the same senior managers are there, but they are sacking CEOs frequently now ( they change them every 3 months now ) so that is improvement.

I expect dramatic improvement ( or bankruptcy. whichever comes first ) in a few years.

And their prices are getting better. They are only a couple of times as expensive as the microsoft products.
And they are improving the stability of their products. True, the documentation still is a menstrual rag, but heck, it will get better too. Just give them a few years.