My post below and a call to a pal has got me thinking. As I am off work this week these are the sorts of things that go through my head as opposed to ensuring contract liability meets error and omission insurance goverage. Today I am thinking about a smarter internet.
For some time, I have bitched and moaned about the essential stupidity of the internet, how you cannot really find things directly, you can only find things indirectly through search engines like Google or through websites whether general like a news service or particular like this blog. What there is not is an index, a card catalogue that contains search identifiers defined by the authors. I have called this a tree of knowledge using the old trunk, branch, twig analogy for the classification of content.
What I am suggesting be created is a keyword tag. A tag is a bit of information written in HTML code that is added to a page or a part of a page on the internet that contains data about the post rather than the substantive content. It is instructions about the unit of content. It is, therefore, meta-data or information about information. Here is an example:
The image to the right has the following within the few words and characters that instruct your computer to display it: title="Kingston, Barrack St. LCBO" alt="Kingston, Barrack St. LCBO". The image to the left does not. If you pass your mouse's cursor over each image and leave it there for a second you will see that the one to the right displays a little title for the image and the other does not. Other forms of instruction I use tell images to center, have a border, align themselves to the right.
What I am suggesting is the creation of a new form of tag called "keyword" that could be added to any text post, within a link to any audio file or within the instructions to display an image. With this tag, authors of any web page can broadcast what they think is important for you to know about content. The keywords could identify the author, particular concepts or general categories. With this tool, a seaching tool would be able to index all postings according to those key words and from them an index of all concepts being posted about. Here are some reaons why the creation of a keyword tag would be good:
- it is voluntary. Just as I can add or not add a title to any photo, I could add or not add keyword information.
- it would allow for the creation of indexing services, web aggregators which would read the keywords and present together content from any source in any form which contains similar keywords.
- it avoids the problem of antipatory classification where I set up ten categories which later I slot what I do into. Often you feel that things go into four categories but you can not recover content based on that unique feature that is captured in the four-ness of your post.
- it is focused on the "power user" allowing them to develop it without hindering the casual user's experience of the internet.
- it would allow for an index of one person's content regardless of where it is posted. As long as the keyword personal identifier is included in a blog post or comment or reply or news item or image or audio file on any webpage, an aggregator would pick it up. Similarly, if I wanted to post something anonymously, I would just leave out that keyword personal identifier.
- Both the expression of keywords and their collection and organization by indexing services would develop organically. The layers of classification of keywords would cross reference naturally over time as would the completity of search elements which could be added to each tag.
- RSS feeds of keyowrds or keyword combinations would allow for customized delivery of the internet's contents as it updated which would be entirely defined by the authors and consurmers of the content, not a middle party such as Google.
I will leave that for now like that. I give it to you free. My gift to pajamastan and those who stoke the coal fired servers it runs upon. Comments?
Comments
'nee - October 13, 2004 3:49 PM
Porn already ruined metatags and proper content indexing. It's a good idea! Current ALT tags should be enough to accomplish what you mean, though, surely... the A tag can already be used as a target without a hyperlink, so perhaps simply using that with keyword= inside of it can be used to delineate subject matter? Anything'll work as long as everybody uses it.
Alan - October 13, 2004 3:58 PM
You tell me but I think you need a new tag. You would not want the keywords showing up on images as they do with alt tags sometimes.
And everyone does not have to use it - only people wanting to would have to, so it is an option but not a requirement.
Alan - October 13, 2004 4:05 PM
On the pr0n thing and meta tags, maybe inclusion in the indexer is by invite or at least human participation confirmation so automated spam is excluded. Maybe indexers would exclude keywords or your rendering of it would include a filtering option to opt out of links to identified spammers or other content you do not want like Google images "safe search" does now. These would be the job of indexers and the best filterers would win the business of the people searching categories or using their RSS feeds.
'nee - October 13, 2004 5:57 PM
Well, my point was that as long as it's adopted universally (by those who want to accomplish the same goal) then it doesn't matter how it's done :) But you're right, perhaps a new tag is necessary. My thought was that it takes a very long time for things like that to get included as standards and then to be understood by every piece software that using an existing tag for a new purpose might be the way to go.
Alan - October 13, 2004 6:17 PM
If a web-based reader were built or insertion into a browser as a mini app - then the standard would follow - not much different than the adoption of RSS except it is much simpler. You would need a group dedicated to the adoption, however, to create the content. That is the real road block.