Prevalence and classification of web page defects

Abstract

PurposeThe purpose of this paper is to provide an update on previous surveys that have looked at the quality of HTML documents on the worldwide web. Previous surveys have indicated that the quality of HTML documents tends to be quite poor, with most documents containing defects.Design/methodology/approachTo determine the extent of this problem, the paper undertook a large‐scale study of HTML document quality among the most popular web sites (approximately 100,000).FindingsThis paper found that the vast majority (over 95 per cent) of web sites did not adhere to the worldwide web consortium standards for HTML.Research limitations/implicationsThis study represents a single investigation over a short timeframe. Hence, ideally the study needs to be replicated in the future to help generalise the findings.Practical implicationsSuch poor quality may jeopardise the security or usability of a web site, making the site's users vulnerable to malware attacks. This poor level of quality has drastic implications for web usability and security.Originality/valueThis new survey undertook a more extensive examination of popular web sites than previous surveys.

Toward characterizing HTML defects on the Web

“…It is interesting to observe that the second most common error and the top warning both refer to the use of deprecated elements and attributes, which means that their use is still quite high among developers. This can attributed to many factors including lack of knowledge regarding the new standard or the indifference regarding HTML validation among other causes …”

“…Ofuonye et al used Alexa.com as a source of 100 000 URLs to find only 3% of valid pages with the analysis per geographical origin showing significant differences. The top problems found were the lack of a DOCTYPE declaration, missing character set, and a conflict between the character set declaration and the meta attribute.…”

“…Currently, we are considering the preliminary indicators displayed in Table mostly based on empirical evaluation and on what was used in previous works. ()…”

“…Nowadays, it is common to find websites that, because of small details, do not comply with the standards and fail to be correctly processed by browsers . Sometimes, developers are just trying to take advantage of browser‐specific features and overlook standards compliance in return of some benefit, whereas at some other times, they just make mistakes, leaving residual defects on the pages.…”

“…Browsers gradually became tolerant to small mistakes, which actually leads developers to further disregard compliance with standards, from the moment that web pages are apparently rendered correctly . The problem is that the web environment is hugely dynamic and browsers are increasingly diverse, being very frequently updated.…”