Hi all,

I'm sure most of us round here have heard of Dino Esposito - *very* well
respected in our industry...

I was reading an article of his in this month's VSJ entitled "Architecture
guidelines for ASP.NET AJAX applications", where he outlines his definition
of the difference between web sites and web applications - you can read it
here: http://www.itarchitect.co.uk/article...lay.asp?id=351 - scroll
right down to the bottom...

I was quite surprised by what he wrote!

Maybe he's not talking specifically about VS.NET 2005 and the website model
vs the web application model because, if he is, then I have to say that I'm
pretty much in total disagreement with him!

I'd be interested to know what others think...

Mark

He isn't talking about the 2 Visual Studio web site or web application
project models, he is talking conceptually. I just finished a post about the
other, if you are interested:

http://petesbloggerama.blogspot.com/...plication.html

--Peter

Site: http://www.eggheadcafe.com
UnBlog: http://petesbloggerama.blogspot.com
Short urls & more: http://ittyurl.net




"Mark Rae" wrote:
[..]

Given the context of the article Dino may be alluding to the breakdown of a
web site into a possible set of service applications. In the AJAX world of
course, any self contained application, or application within a larger web
site exposed correctly can offer service and thus be a web application to
which an AJAX consumer can subscribe. so site offer interaction where
applications offer service. In this more generic context you dont need to
know about the whole web site to use the applications within it, and so
applications and sites can probably be seen as very different things. Thats
my philisophical take on it!

Regards

John Timney (MVP)
http://www.johntimney.com
http://www.johntimney.com/blog


"Mark Rae" <mark> wrote in message
news:4132
[..]

"Peter Bromberg [C# MVP]" <pbromberg> wrote in
message news:47d8

> He isn't talking about the 2 Visual Studio web site or web application
> project models, he is talking conceptually.

That's what I was wondering... Problem is, he doesn't make that particularly
clear, especially as the text is part of a larger article relating to the
architecture of ASP.NET apps...

> I just finished a post about the other, if you are interested:

Couldn't have put it better myself!

He's definitely NOT talking about Visual Studio.Net's programming models for
ASP.Net web applications. In fact, he's not tlaking about Visual Studio.Net
at all. He's talking about the general architectural concept of a "classic"
or "thin-client" web application versus one Ajax model, which employs a
thick client and a REST (REpresentational State Transfer) back end.

On May 7, 1:34 pm, "Kevin Spencer" <unclechut> wrote:
[..]
>>
>>
>>
> - Show quoted text -

Yup, he is not talking about VS.NET at all... however, his words are
strange: "A Web application is somewhat simpler. It is a relatively
small application implemented through one or just a few Web pages...
Gmail and GoogleMaps are good examples of ... Web applications". Why
is it simpler and simpler than that? I don't think this can be
compared by a number of the pages.

"Alexey Smirnov" <alexey.smirnov> wrote in message
news:9740

> Yup, he is not talking about VS.NET at all... however, his words are
> strange: "A Web application is somewhat simpler. It is a relatively
> small application implemented through one or just a few Web pages...
> Gmail and GoogleMaps are good examples of ... Web applications". Why
> is it simpler and simpler than that? I don't think this can be
> compared by a number of the pages.

That's what puzzled me too...

I disagree with Dino, but for different reasons than you do (mine is
partially focused on semantics). He is not talking about ASP.NET application
templates (the website versus web application models adopted by Microsoft).

A web site is a collection of pages that sit under a single domain or
subdomain. In other words, everything at www.mycompany.com is part of that
web site. Dino restricts web site to only be static content, which is an
unnecessary restriction of the definition. But, it is a definition that can
work with his examples.

A web application is a web site that includes code to interact with the
user. In most cases, perhaps all, session state is important. In most, you
have clumps of grouped functionality. Dino takes this farther to restrict
the definition to web apps that use a Model-View-Controlller (MVC) or
Model-View-Presenter (MVP) pattern. I also find this to be an unnecessary
restriction as one can use a model where each page presents a single view
(ala, classic ASP), which would make our simple ASP.NET site a web site, but
a complex ASP.NET site a web application?

Summary:
Web site = everything under a single domain or subdomain (if a company has
many domains)
Web application = subset of web site that includes interactive content
(application)

In his world, he is leaving the possibility that you have multiple
"applications" per site. I can see his point, but web applications are
encapsulated in websites.

A web site is a collection of resources (pages, interactive pages, full
applications, PDf files, etc.) that reside under a domain (or perhaps
subdomain, ala www.microsoft.com, msdn.microsoft.com, et al). A web
application can be an interactive web site or a part of a web site.
(correcting my initial reply :-( )

"Cowboy (Gregory A. Beamer)" <NoSpamMgbworld> wrote in
message news:1ef6

> He is not talking about ASP.NET application templates (the website versus
> web application models adopted by Microsoft).

That's what I wondered originally... :-)

However, the text in question is included as part of a much larger article
relating to the architecture of ASP.NET apps which I read originally in this
month's VSJ, which is why I found it a bit strange at first glance...

One correction to add here to my statements:

A web site is a grouping of pages (correct). One typeof web application is
an interactive web site. But, a single web site might have n applications
(potential one to many). Dino is restricting this to his AJAX plus REST
version of web apps, so the definition fits his example, but not all sites
or apps.

Think of a "web site" as composed of several "web applications",
and the meaning he intended becomes clearer.

What he wrote doesn't apply to VS nor to ISP hosting, for obvious reasons.

The writing is a bit fuzzy ( uncharacteristic in Dino ) because a web site doesn't
necessarily contain more than 1 application, so a web site *could* consist of just
a few dynamic pages with a database backend.

And...you are both right: whether a web site is a "web site"
isn't determined by the number of pages hosted in it.

I'd say it's whether there's a single corporate umbrella handled in its pages.

You could have business divisions handled by different parts of a corporate web site.
I'd call those "applications" regardless of the number of pages.




Juan
====
"Mark Rae" <mark> escribió en el mensaje
news:4872
[..]

It's not surprising that you found his article confusing. Dino is one of the more prolific writers but unfortunately, also one of
the worst writers, in my opinion.




"Mark Rae" <mark> wrote in message news:4132
[..]

re:
> one of the worst writers, in my opinion.

Your opinion is unfounded, in my opinion.
Dino is not only prolific, but very clear in his writing.

This is the first time I've ever seen relatively unclear writing from him.
Maybe you could support your opinion with other examples of "bad writing" by him ?




Juan
====
"Jon Paal [MSMD]" <Jon[ nospam ]Paal @ everywhere dot com> escribió en el mensaje
news:3996
[..]

Look troll, we can't debate an opinion, it would be like debating a color preferrence.

It's my opinion, if you don't like it ignore it.

Besides, I've told you before, don't reply to my posts and yet you persist like a sick-minded stalker.




"Juan T. Llibre" <nobody> wrote in message news:4132
[..]