Web 2.0 and PHP

jonki

How compatible is Web2.0 with PHP? I understand that using PHP and Ajax or javascript I can create Web 2 webpages. Am I right? I have a couple of quotes from external web companies to create full Web2.0 pages of all my sites but they dont work in PHP so would have to start from scratch. If I hired the correct people, could I slowly update my current PHP site, slowly into a full Web2.0 site just using javascript?

This may be a stupid question but I have read up a lot on Web2.0 but havent really found any sites which can answer me these questions without me going into lots of detail.

Many thanks

Jon

NogDog

Inasmuch as "Web2.0" has a rather vague definition, it might be best to tell us how you define it before we try to discuss how it works with PHP. Generally speaking, however, I see no incompatibility between them. In fact, most of the functionality associated with Web2.0 requires some sort of server-side programming, of which PHP is certainly one of the most commonly used tools.

Kudose

Web 2.0 is just a buzzword basically meaning things are dynamic and user interactive.

What's funny, to me at least, is that the technology has been around since the late 90's.

If you make a form using XHTML and post to a PHP script using JavaScript and update your form with the response text via JavaScript, you've just made a "Web 2.0" form.

TeraTask

jonki wrote:
to create full Web2.0 pages

Whoever defined this requirement doesn't understand the term "Web 2.0". In simple terms, Web 2.0 means that when a link is clicked, you don't necessarily have to go to another URL. In practice, it has been implicitly defined to be any dynamic javascript which looks cool, but as previously noted, such features have been around for quite some time. The condition "full" must be more completely defined for you to have any chance of fulfilling the requirement. Let me provide some questions which may help:

1) What is meant by "full"?
2) Are fancy page transitions required when new content is loaded?
3) Do sections of the page need to be able to be arbitrarily moved around (i.e. drag-and-drop) and the resulting locations saved on a user-by-user basis?
4) Should text fields auto-suggest possible entries based on each key entered by the user? If so, can it be every third, fourth, fifth, ... key entered by the user (NOTE: the more keys allowed, the less work required by the server).
5) Which browsers should be accommodated. "All" is not acceptable as there are lots and lots of browsers. I usually support the last 2 versions of IE, the latest version of Firefox, and the latest version of Opera. Customers pay more for the substantially-less-popular browsers as relatively few of them support the full features necessary to achieve certain tasks.
....

It could really go on forever. You should try to find out how the client wants the end site to function and then do whatever coding is necessary to achieve the achievable. Clients should be able to say, at most, that they want a site to use the latest technologies. Depending on the requirements of the site, you will then need to use the techniques which are most appropriate, efficient, and reliable to achieve the end task.

And, inform them that not all users have javascript enabled (a necessary precondition for support of Ajax [the true tech behind Web 2.0] on a wide variety of browsers), so a contingency must be in place for those users (think about those with screen readers, for example) who do not have javascript enabled.

Weedpacket

TeraTask wrote:
5) Which browsers should be accommodated.

Depending on the intended audience for a given page is, there is at least one other client that should be accommodated. Googlebot doesn't do JavaScript; so if this is a public-facing page (as opposed to a members-only/user/private page) intended for presenting content (as opposed to configuration/application UI) then it should remain legible and functional even if all the JavaScript has been removed. For those sorts of pages, keeping search engines happy isn't really a "contingency" 🙂

TeraTask

1) I kinda doubt you trademarked (or could) "Custom User Title", but that is interesting.

2) More on topic, I have written an article on my site (go to my username with dot com and view the articles section for details) which allows for Ajax-Search-Engine-friendly websites. I would consider such support as supporting Lynx and any other text-only browsers as well as search enignes.

leatherback

TeraTask wrote:
2) More on topic, I have written an article on my site (go to my username with dot com and view the articles section for details) which allows for Ajax-Search-Engine-friendly websites. I would consider such support as supporting Lynx and any other text-only browsers as well as search enignes.

Nicely written article, although hardly worldshocking 😃

Jack_McSlay

Weedpacket wrote:
Depending on the intended audience for a given page is, there is at least one other client that should be accommodated. Googlebot doesn't do JavaScript; so if this is a public-facing page (as opposed to a members-only/user/private page) intended for presenting content (as opposed to configuration/application UI) then it should remain legible and functional even if all the JavaScript has been removed. For those sorts of pages, keeping search engines happy isn't really a "contingency" 🙂

I've found out that search engines usually give out their URLs in the browser signature, so a

$isbot = strpos($_SERVER['HTTP_USER_AGENT'],'http://') !== false;

should catch most search bots.

Still, a more failsafe trick is to serve the first page as a page readable without javascript, and use javascript to set a cookie telling the browser is javascript-enabled.

As for text-based browsers, I've never heard of a single person who uses them regularly, so it's agueable if it's even worth the trouble supporting them.

bradgrafelman

Jack McSlay wrote:
As for text-based browsers, I've never heard of a single person who uses them regularly, so it's agueable if it's even worth the trouble supporting them.

I think the idea of looking at your website in a text-only browser (e.g. Lynx) gives you an idea what a search engine will see. In other words, if you can't figure out what content is on the website using Lynx, then Google most likely can't either.

TeraTask

I honestly don't install Lynx. I just install the developers bar into Firefox and then check the boxes to disable javascript, styles, etc when testing. To each her/his own, but I wanted to share the method.

leatherback wrote:
Nicely written article, although hardly worldshocking 😃

That was my take on those famous finger print works done by some famous artist long ago. When I said something like that, the teacher pointed out that while it may not have been the best thing in the world, he was the first to do it.

I did a lot of research before putting my head to coming up with a solution and found such advice no where else on the internet. I posted the article to help others find the solution I had worked out. Not, to shock the world.

leatherback

😃 opinions might differ 😃

I think that if you start looking on this forum, you'll find that the main advice on using AJAX for any function is to make sure you have a java-script independent version running, and put the AJAX on top. Which, in summary, is what your paper comes down to.

But: It is good to see someone write it up in a clear consize manner.

TeraTask

Javascript should always be "put on top" if one does it right. Doesn't mean that people figure out how. Check around the web and see how others do it -- the javascript is still on top, but they manipulate the anchor. I leave the anchor alone.

Oh, and let me say, Thank you. I appreciate the compliment on the article.

Weedpacket

TeraTask wrote:
I leave the anchor alone.

Something like the suggested method in "Unobtrusive JavaScript"; e.g., http://www.onlinetools.org/articles/unobtrusivejavascript/chapter3.html then; leave the link as <a href="http://www.example.com/document/page/2">Page 2</a> and then the onLoad handler can roll over the document adding onClick handlers as appropriate.

I did this for an image search page recently after reading the above article; all of the links (pagination, results-per-page, keywords, sorting, and other criteria) functioned both as pure HTML links which reloaded the page, and when the JavaScript was loaded, as event triggers that refreshed the results display and updated the rendering of all the controls as appropriate.

TeraTask

@ That dawned on me as I was posting to this thread too. I'll be making that change when I get back in there and do some coding.

Thanks!