HTML Must Die!
Since my first Apple (a very old Powerbook100) the surfing in internet had a name: HTML.
Many years have passed since then, but the HTML remains, being the most widely used language in the world (gimme anyone with a computer that have NEVER surfed on a web page…)
Even if the HTML has evolved in DHTML, saturated with Javascript, automatic postback, it remains a STATIC “thing”.
Do you wand any dynamic contents? You can Cheat, you can make a big bluff, preloading everything you need, but the HTML will remain Static forever
It seems to me that HTML have played the role of the tortoise in those years, while the rest of the informatic world has run like hell…
The problem is that the html has its own stronghold, without any good proposal for a new standard. No more Internet Explorer, Firefox or Opera, but a brand new generation of www browsers should see the light, to bring more advantages for everyone.
I think that HTML must be REPLACED to give space to new technologies that will be able to get its inheritance.
Now the requirement to have exchangeable, dynamic data is a requirement in most of the softwares, but where is html?
Why do we need HTML anymore?
First of all I have to say that html is EASY. Also a child can create a good web page with a text editor. This is a Big Advantage, obviously, and I have to say that it is also very flexible with the CSS (even if the css is harder to be used well).
But this does not move the point of my interest.
The Dynamical data exchange.
Nowadays we have to move from the server to the client (any browser) the whole page, completely written and encoded. This is not good, because lots of informations are moved every page refresh: why don’t we try to code a new way of defining a web site, where the informations about the visualization are passed one time only and reused continuously?
Why couldn’t we modify our “HyperText Markup Language” in a more complex, fast and reliable “Dynamic Web Content Management”, permitting the WWW data to become a new hare?
Let’s take, as example, a page in which you can load a list of “Manufacturer” and for everyone of them, all their “Products“.
- You can choose to load the complete list of “Manufacturers” and the whoule list of the “Products” of every producer, sending those formatted data to the Browser and let javascript filters the “Products” for the selected “Manufacturer“.
This mean a large amount of data passed the first time, but then we wouldn’t have to wait any longer (except for the Javascritp functionality). If you are managing a small amount of data this can be good, but if the data have a large amount of “Products” for every “Manufacturer“, this cheat will not be good any longer, first of all because of the high time needed to receive all the Formatted Data, and second of all because the Javascript will get a lot of time to render the sublist. - You can choose to use another method: you can load the “Manufacturers” list and, when choosing one of them, recreating a new page with all the “Products” from that “Manufacturer“. This will mean a shorter time to load the first page, but more than one page with “already HTML formatted data“.
With this method you can reduce the time to load every page, but you have to require more pages to get to the same results as in the first example.
Why do i pointed out on the “HTML Formatted Data”?
Well, because it uses string data (”“) instead of using Objects.I will not start a disquisition on objects right now (I hope you know what they “are”, if you do not immagine them as “entities” that have specific properties and functions that identifies them clearly).
Passing object instead of html code will help to build a structure based on “abstract” information that will be joined with a “visual” layer directly on the client computer.
I also want to tell that this way of sending data, In My Humble Opinion, could reduce the amount of data passed through the net, resulting in a higher speed for applications (or even websites) to be loaded.
In my mind I can see a structure that exchanges different types of data.
- Visual Information
- Data Information
- Event Information
The main reason for my cursade against html is that the method I have in mind and that I tried to show you shortly could let the www become a real “Dynamic” environment, moving the integration between the layout and the data from the server to the client.
Let’s immagine a data flow that delivers the information needed for the rendering of the first page, together with some data (to be written inside). When the information are completely delivered to the client, the server could send the data for the other “New Web Pages” while the “New Web Browser” shows the first page to the user, in background. This could cut down the time needed for every other page loading.
Then immagine, like in the previous example, two different selections: “Manufacturers” and “Products“.
When the page is loaded the client requires only the “Manufacturers (Object)” list, merging it in the layout. When a single manufacturer is choosen the client send the information to the server, that return only the choosed manufacturer’s “Produts (Object)“. Another time the browser will merge the informations, without even performing a page refresh, because the new standards should be more like a Form (Windows Form) than a html page (in the functinalities, not in the eyesight…)
Wouldn’t it be good?
We should encourage a new vision of the “www”, moving the focus from the simplicity to the dynamicity of the layout and contents.
Come on, people, tell me what do you think about it!
1 Comment