Site Design

[stryder]

You will eventually notice a slight change to the site design between the old overlay method for identifying new/old/locked and linked forums and a new method.  I'm not sure which is better, or for that matter if there is perhaps an even better method round the corner.. in any event feel free to post what you think.

Original/New:

[Image: forum_icon_sprite.png]

[Image: forum_icon_sprite2.png]

Personally I always found the tiny icons not particularly defined (fuzzy)

Additionally you'll notice the QR Code:

[Image: scivillage_qrcode.png]

While it's likely that nobody will use it (it only contains the URL for the website) it's still something that will likely find itself in the image cache of search engines and perhaps used to float the URL out there into the real world (For instance printing it on a balloon and then filling it with helium and releasing it, or putting it on a piece of paper and putting it in a bottle to be thrown into the sea as the tide goes out etc)  In any event it exist for whatever reason might popup in the future.

[stryder]

You might have noticed some additions to the design pop-up as I try to work through it. I can't say for certain if everyone is actually seeing how it's suppose to be shown though, so some might see errors where I and others do not.

For instance I'm currently working on a new way of outputting certain information like timestamps, navigation words etc. The main reason for this is because search engines spider the site and then attempt to mark-up the sites keywords by their weight (how many times they are applied) the problem is that some words are used far more commonly than they should be (for instance "login" or "register") and that can lead to some mishaps when people search.

The only search engine/spider that currently uses a technique to remove text from scraping from within a document (as opposed to the whole document) appears to be Yahoo (perhaps Bing too)

In any event the following is what I've been working on:

Code:

<style>
.x:before{
content:attr(title);
display:inline;
}
<style>
<span class="x" title="hello world"></span>

The result is that the content of the alt attribute is actually written where the Span exists in HTML. It means that a search engine when looking for a value in span can potential ignore it, meaning that "hello world" in this instance doesn't get weighed as a keyword.

However there is a potential problem with this, search engines can actually strip data from title tags to work out keywords (so it kind of counters it) what it would suggest is that in search engine etiquette it would make sense that if a tag doesn't have a value but does contain an title, it should actually be ignored. The title should only be considered when the value contains some data.

An example of it's usage currently is that all the Date stamps on the Forum index page are now created in this manner. (If you can't see dates, then there's a bug)