Verwijderd schreef op zondag 10 januari 2010 @ 14:29:
Ik denk dat in dit geval een document uit 2007 anno 2010 niet zo gek veel waarde meer heeft

Onze vrienden van google veranderen toch wel met enige regelmaat de ranking logica. De trend op dit moment is toch dat voornamelijk de inhoud de meest waarde heeft (en natuurlijk nog wat andere factoren). Alle beetjes helpen natuurlijk wel.
Bullshit, het hele web is erop ingericht om relatief weinig waarde te hechten aan GET-parameters, juist omdat ze bedoeld zijn voor repeatable en quotables actions zoals searches, denk aan /search.php?q=pizza, waar Google zelf ook stijf van staat als zoekmachine. Dat is waar GET-parameters voor bedoeld zijn en hoe ze geindexeerd woorden, als
parameters voor een pagina. Op basis van die logica is het in principe terecht om te stellen dat een site die enkel werkt met 1 pagina en parametrized URL's (
voorbeeldje) maar 1 pagina heeft en daarop veel input-afhankelijke variaties.
Wat jij even niet wil snappen is dat URL's wel degelijk bedoeld zijn om human-readable te zijn. De gemiddelde gebruiker stelt het extreem op prijs dat ie in z'n mail/IRC/MSN/whatever URL's aantreft als
http://tweakers.net/nieuws/64839/bioshock-ontwikkelaar-2k-boston-wordt-weer-irrational-games.html en niet zoiets als
http://tweakers.net/index.php?type=nieuws&id=64839. Niet alleen is de 2e veel minder verbose, hij draagt ook niet duidelijk uit dat het een unieke pagina is, en hij is niet
'hackable', want van de eerste mag je stukjes afhakken tot bijv.
http://tweakers.net/nieuws, om een logisch overzicht te krijgen van al het nieuws. Hackability is ook een belangrijke reden voor URL-rewriting.
Ik weet ook wel dat je koppig bent en graag je eigen mening als wet verkondigd, maar kom op, in deze heb je opponenten als
Jakob Nielsen tegenover je:
Edward Cutrell and Zhiwei Guan from Microsoft Research have conducted an eyetracking study of search engine use (warning: PDF) that found that people spend 24% of their gaze time looking at the URLs in the search results.
MSR used Microsoft's own search engine (fair enough), but their results match what we found in our eyetracking research which included the current market leader as well as the #2 search engine in addition to MSN.
Users have evolved a firm model of search behavior which they apply across search engines, which is why it's probably a lost cause to make a non-standard search user interface.
We found that searchers are particularly interested in the URL when they are assessing the credibility of a destination. If the URL looks like garbage, people are less likely to click on that search hit. On the other hand, if the URL looks like the page will address the user's question, they are more likely to click.
Maar dit is helaas ook een addendum uit 2007 op een tekst die reeds sinds 1999 als 'bijbel' beschouwd wordt, dus vast verouderd...