Re: PHP on .NET DLR

On Sat, Aug 11, 2007 at 05:14:44PM -0400, Mike Schinkel wrote:
> 
> Antipodal reaction?  LOL!  You kill me!  Different managers with different
> view of priorities, that's all! 

 "Antipodes" is shakespeare. I have always wanted use it somewhere. :-) Nah. There is a real conspiracy here. You just have to delve deeper. You are being so naive.

> That statement tells me you don't even realize why the web technology is so
> resilient and has scaled so well. I only understand it because I recently
> spend six month diving deep into old mailing list discussions and reading
> specs, and I came out a believer.  Loose coupling that can accomdate failure
> is key to the web, and strings provide that. For more local apps, structure
> is okay. But even for local apps the lack of the equivalent of a UNIVERSAL
> resource locator for stateless resources is a serious downside to that which
> you hold in high regard.  
> 
> But certainly it would take me years of repeating other people's arguments
> to get the point across. May I suggest you read RESTful Web Services [1] and
> then we could have this discussion?

 Yeah. But there exists also a radically different approach in the Desktop, which is also EQUALLY succesfull, and the real reason why web is hot is that customer doesn't need to download anything. Anyway, let me check the link and I will tell you if it makes any sense.


> 
> > silverlight will succeed or not is irrelevant; it can't hurt 
> > anyone to have PHP on .NET nevertheless.
> 
> It can distract attention from where it might be better utilized.  FYI, I'm
> not against this I just have a natural tendancy to provide a counterbalance
> when I see other people being more zealous than pragmatic.

 But it is already there. Phalanger is already there. The only thing is needed: Zend has to get MS adopt it. If Phalanger didnt' exist, I wouldn't bother. I have many many times repeated that the only reason why my request cannot be deemed unreasonable is that Phalanger already exists in a completed state, and I am currently using it to build applications, but the project looks abandoned, and nobody is answering any of my queries.

 


> 
> >  OK, silverlight fails miserably. So what? PHP on .NET will 
> > still empower millions of PHP developers to explore new 
> > avenues, and come into contact with the new programming ideas 
> > that .NET library has. It can only be good.
> 
> Maybe. Or maybe they just code in PHP classic and go about their business.
> 
> Listen, I'm not against this. I'm just saying "put your money where your
> mouth is" and get coding.

 Code is available. Only thing needed is the official stamp from MS.


> 
> > RSS is not a language. 
> 
> Hmm. According to XML.COM, RSS is "a portal content language."   The reality
> is a computer language is an abstraction layer that allows the user to
> encode information for processing and/or representation.  Your point is
> quibbling a nits.  RSS is a declaritive as opposed to procedural language
> and the simplicity analogy still applies.

 I would say you are the one who is nibbling at the quits. We are talking of building applications. If it doesn't have while loop, for loop, and if then, then it is not a language. Sorry.

> 
> Never said there way.  My point about independant components actually
> supports that concept.
> 
> > Complexity is the key. And complexity exists so that it can 
> > hide itself from the end user. That's programming. You make 
> > your programming incredibly complex so that your user doesn't 
> > have experience it.
> 
> You "make them complex?"  Do you really hear what you are saying?!?  No, you
> factor the complexity out so that you can handle the complexity without it
> causing you to make mistakes.  

  Complexity is unavoidable. If it was simple, then there is nothing has to do. That's a tautology actually. If it is simple, then it would have been already done by some idiot. The only tasks that are left are the complex ones. So complexity will ultimately exist at some place. It is your decistion to have it in code or pass it on to the user. My preference is to have it in code, and make it easy for the user.


> "Not meant" for writing applications?  What inside information do you have
> my dear friend?  Have you travelled to Redmond and met with the people on
> the Powershell team and discussed Powershell with them?!?!?  (like I
> have...)

 Can you write Silverlight in powershell? In that case, I think I will stop wasting time here, and learn Powershell better. Powershell is pure C with $s.

> 
> > But funnily enough, I love powershell, because it is pure C#. 
> 
> According to the book written by the PowerShell team lead, Powershell is
> part Perl and part Python. It is not part C#.  And it doesn't have those
> infernal semi-colons. '-)

 It HAS semi-colons.  It is 100% C, with $ prefixed to variables.

> > Python doesn't even allow you to embed variables in strings. 
> > How insane can it get? 
> 
> Here's how insane: You can take one feature from a language and claim it
> discredits an entire language. Now THAT is insane.  The more you type, the
> less credible you sound.

 When there are competing languages, every feature counts. Why should I bother with python, when I can get the same and MORE from Ruby? If python alone existed, and the other choice was perl, yes, you have a point. Otherwise, no, why put up with nonsense, when you can get the same and more from somewhere else?

 And that's exactly what happened. Before Ruby, python, despite its flagrant disregard for the wellness of the programmer, did well, because the other choice was even worse. But not anymore, and for that I am thankful.



> 
> > And you need to add 'self' to every 
> > method in a class. 
> 
> That's a design decision that many people agree with, myself included.

 Redundancy is terrible. Why should you want to clutter your code with a self everywhere? If you are in VIM, you can color $this, or self separately, so there is absolutely no chance of making a mistake. Also, after sometime you automatically understand when you are inside a class and when you are not. You don't need a flagpost of a 'self' cluttering your code to tell you that. That's a bad decision, and implying that I am not capable of realizing when I am inside an object, and when I am not.



> >  Why stop there. Why not take the next step. As I said, for 
> > backend I will be using PHP/IIS, but what will I do for the 
> > silverlight frontend?
> 
> Nothing is stopping you. Go for it.  (we are full circle now.)

 Get Zend to get MS get the old programmers back to PHalanger. (I know 3 gets).

> > 
> >  So now the next step, what can we do to make this happen?
> 
> Recruite a team of crack developers who want the same and pick of the
> Phlanger project.  Heck, go talk to Miguel at Novell and ask if he can
> provide you some guidance.
> 

 The code is there. The only thing lacking is the official stamp. Both from the PHP community and from MS. Tomas has done an outstanding piece of work, but it got scuttled by MS. Why do I have to repeat this again and again. the code is already there. If it didn't, I am merely trying count the clouds. But that's not so.

 Thanks.