Chapter 12: Semantic Web

The time to go deeper has come.

A proposal for the web

I have seen many proposals before, but in my opinion they do not entirely work. I'm going to say something that a lot of people might not like:

Once again I’m sorry, but that’s the way it is. HTML has never been a good semantic language, although it's okay for presenting content. A lot has been done in order to separate the content from the presentation, but I think that there is still a long way to go. And I don't know if it is being made in the right direction.

What do we have right now?

Well, there is HTML for the content. Then we have CSS for the presentation. Mix, shake well, and we get a fantastic website. We also have large current search engines, from which we generally get good results. Not always the ones we want, but that's inevitable.

But back to HTML and CSS. Is it really semantic? I don’t think so. HTML4 was obviously not, and we have now evolved into HTML5 but, what have we gained? Much in terms of standardization and language evolution, but make no mistake: HTML is still based on the presentation.

What is my proposal? Are you really asking?

Alright, it is clear, STxT. And how do we make the semantic web with STxT?

I don't know, although I imagine it’s one of the best ways of doing it nowadays :-D

What did you expect? Did you think I would have done everything for you already? If we get somewhere with STxT, we will see. My visionary powers are not that advanced :-)

The following sections are ideas, things I have thought about, but there is no specific proposal. There are many scenarios in which we can imagine this new technology, and many ways of applying it. I don’t even know now, but I can see its potential. I hope others also see it and know how to apply it.

Ideas

Replace HTML5!!

But that is crazy!! We have spent years evolving, making browsers,... how could we do something like this!!

That is crazy >:-O

I know, but let me show you my vision. This goes for computer techs, and people who work with HTML.

With STxT everything can change, it is a new world, a new way of thinking. Let’s not close our minds to anything.

Clear your mind. There is nothing there now. Nothing exists yet. We are in 1985. We have never heard of SGML, HTML, hyperlinks, or CSS. We have an acquaintance with a PC and they use the command line to do things.

Someone tells us that a new language has been invented. It is called STxT and it serves for expressing ideas and displaying text.

We now have a vision of the future, we are in 2020, and we see a web, but we don’t know anything about computing. The web looks very much to the one we have nowadays, but everything is structured with STxT.

You ask yourselves... How is it made? What is it like inside?

...

...

...

...

What have you seen?

Well, as you cannot answer I will tell you what I have seen. It can help you or it may be nonsense, but it is my vision.

Web pages are STxT pages, with each web and portal defining its own structure.

This lets us see the web as a distinct content set of pages. For example, for a cooking web:

/recipes/*.stxt
/tricks/*.stxt
/diets/*.stxt

Apart from this, every web portal has a series of pages that give us information on how to format or present the text. There is a group of people who did this and did a great job. They did it in the W3C.

They made STxT documents that allowed this, and all webs used them, thanks to their popularity. Besides, as there were fantastic STxT viewers, everything was able to be integrated without problems, and many viewers were created.

The pages that needed to have a website were quite a few. The main ones were:

robots.stxt
sitemap.stxt
styles.stxt
/styles/*.stxt

These pages allowed us to get an idea of the portal content, while at the same time we applied styles to the whole. The best part about styles is that they were automatically applied depending on the type of document, and it was also very easy to make new styles. A document which followed a known namespace and easy to use was created: w3c.org/css.stxt

Thus, the pages were very easy to create. With a simple STxT editor, a page with perfect content was built. A designer created a fantastic style for it, and finally a programmer gave it a dynamic behavior creating a js file for each type of page (a scripting language for providing behavior to these pages).

Thus, for example, our cooking website would be like this:

robots.stxt
sitemap.stxt
styles.stxt

/styles/receta.stxt
/styles/trucos.stxt
/styles/dietas.stxt

/dynamic/receta.js
/dynamic/trucos.js
/dynamic/dietas.js

/images/*.png
/images/*.gif
...

/recipes/recipe1.stxt
/recipes/recipe2.stxt
/recipes/recipe3.stxt
...

/tricks/trick1.stxt
/tricks/trick2.stxt
/tricks/trick3.stxt
...


...
...
...

Well, that’s it. Excuse the scheme simplicity, but a hypnosis session cannot be stretched out any more :-) I don't know if this will awaken you imagination, but I'd like to at least make you think. Not everything is fine just as it is. There is always room for improvement. For doing things differently. And we are computer techs. We have to test things. And experiment. And make mistakes. It is the only way forward.

Search engines

I believe that the search engines and browsers in general would increase their searching capabilities and their capability of indexing contents knowing what the text is about. In addition, it would already be nested and each part would have a specific and well-defined meaning.

It’s possible that famous STxT templates could arise, and a page could follow one of them in order to display certain content. This would make indexing even easier, as it would unify certain aspects.

The transition

What should the transition be like? It should be easy and transparent. Otherwise, there will be no transition. The first thing to do would be to define the new way of seeing documents with STxT, and the new CSS. By the way, it would be a good moment for making the CSS and having a first version. Perhaps the browser renders would finally treat them all the samel!

From here on we can foresee a period of page coexistence. There will be good old HTML pages along with STxT pages. I imagine also a good moment for integrating and unifying a "lightweight language" for the content. I'd like for W3C to become interested in this whole process. Although it would be a very alternative front to everything that is being done now. But it should not be traumatic either. We can think of STxT as another extension to be considered by the browsers, with a special way of displaying. The display would be very similar to XML when there are no css or anything standard, and as the web incorporated them, the appearance would improve.

We can even consider the coexistence of HTML pages with a more structured STxT counterpart. The HTML would contain metatags that would tell the search engines that it has an associated STxT, as the indexing using this file would be more appropriate.

We must not forget that the aim of all this is to achieve the semantic web in a less traumatic way, which is easy for everybody, not just for super-computer techs.

SWB: Semantic Web Builder

I couldn’t finish this chapter without speaking of SWB. This is the first STxT application that we have made. Well, I know that it is the first one because it has been made for publishing this web, and before there was nothing else :-D

We could also say that it is the first website to show the content not only in HTML but also in STxT.

I can say that the experience of using STxT has been really impressive.

I've been programming for the web for years, and I had never before felt so comfortable. Even without editing tools, the book creation process has been simply a pleasure. I encourage everyone to try it.

It's worthwhile ;-)