How do we browse those things that are non-browsable.
Deep/Invisible/Hidden/Dark Web
Centres around the notion of searchability - it is not indexed by search engines so it is unlikely to be found.
So, how do we find things on the dark web? Traditional browsing, social bookmarking, word of mouth, blogs.
Libraries as analogy. They are highly organised and can be browsed or found via an OPAC. Closed-stack with no OPAC is of little use to the public. Each website is like a library - with "shallow web" - that is publicly accessible with advertised existence - and "deep web" - not open/known to public this does not mean concealed, but simply obscure.
Organisation leads to usability because structure enables automation. Web indexes create a highly organised structure external to the content - this makes finding your stuff much quicker.
Why is it Hidden?
Intentionally - may require log-in, reputation may be good without indexing of deep content, may be technically difficult to provide the data needed.
Accidentally - laziness, carelessness, indifference, lack of knowledge.
Data and Metadata
Increasing level of abstraction but this goes along with increasing utility. Adding good quality metadata means you can make your data more findable. Metadata are usually concise, precise, well-specified, and uniform across a variety of content. Web authors can provide their own metadata (self-cataloguing).
Complement any search facility with a browsing interface. Avoid JavaScript - bots don't like them. Keep URLs distinctive (URL should determine content). Flickr as both searchable and browsable.
HTML IMG element requires ALT attribute (improves accessibility to vision-impaired users and enables images to be indexed by keyword).
Remember: not only humans are looking at your site-love your robot friends! (Or Clamps will initiate the clampage...)
Showing posts with label metadata. Show all posts
Showing posts with label metadata. Show all posts
15.1.10
Webs of Miasmic Joy (Class 5 Cont. ) - Chris Edwards
How the Web Works
Client/Server architecture: the server hosts documents from which clients request documents via a URL. Once a document has been received by client, the server disconnects [right about now I am struggling to push Celine Dion back into her wee box...].
URI-uniform resource identifiier. URL (locator) is a subset of URI.
URLs are good when they are: descriptive and meaningful (makes it easier for humans to point client in right direction), stable, and technology-independent (if tech changes there should not be need to change the URL).
URL Shortening Services
Having an informative URL is useful for various purposes (e.g. searching). But certain media handle shorter URLs better. A long URL could be changed by software - particularly for longer addresses. Short URL points to a re-direction server which re-routes connection to desired location. Yet such services represent increased chance of link-rot since there are a smaller number of combinations to choose from. Also providers can disappear. Potential security risk in that the link does not point to the target domain - we could end up at a site of nastiness - perhaps called DoomBotCentral (or even DoomWormCentral, DoomPhishingCentral etc...)
HTML
Elements and tags - computer geeks adore trees (is this perhaps a sign they are really repressed hippies?) but they like them upside down with root at top (perhaps a sign that they really don't get outside enough). Elements - a paragraph, heading, image, link - "any discrete unit of meaningful, discrete information." Elements are expressed through tags. Metadata in the head + title (not actually part of the page). HTML - dictates the logical structure of your document whilst style-sheets tell you how this logical is viewed.
What about Web 2.0?
Existing internet and WWW infrastructure - simply the way in which people are using the web structures to prosume info. Infrastructure was not designed to be dynamic but a repository of static documents.
Client/Server architecture: the server hosts documents from which clients request documents via a URL. Once a document has been received by client, the server disconnects [right about now I am struggling to push Celine Dion back into her wee box...].
URI-uniform resource identifiier. URL (locator) is a subset of URI.
URLs are good when they are: descriptive and meaningful (makes it easier for humans to point client in right direction), stable, and technology-independent (if tech changes there should not be need to change the URL).
URL Shortening Services
Having an informative URL is useful for various purposes (e.g. searching). But certain media handle shorter URLs better. A long URL could be changed by software - particularly for longer addresses. Short URL points to a re-direction server which re-routes connection to desired location. Yet such services represent increased chance of link-rot since there are a smaller number of combinations to choose from. Also providers can disappear. Potential security risk in that the link does not point to the target domain - we could end up at a site of nastiness - perhaps called DoomBotCentral (or even DoomWormCentral, DoomPhishingCentral etc...)
HTML
Elements and tags - computer geeks adore trees (is this perhaps a sign they are really repressed hippies?) but they like them upside down with root at top (perhaps a sign that they really don't get outside enough). Elements - a paragraph, heading, image, link - "any discrete unit of meaningful, discrete information." Elements are expressed through tags. Metadata in the head + title (not actually part of the page). HTML - dictates the logical structure of your document whilst style-sheets tell you how this logical is viewed.
What about Web 2.0?
Existing internet and WWW infrastructure - simply the way in which people are using the web structures to prosume info. Infrastructure was not designed to be dynamic but a repository of static documents.
Subscribe to:
Posts (Atom)