And the series continues…
This time it’s Chrome. Chrome is an excellent browser, but some dialogs are weird… And some are not taking no for an answer.
The dialog below asks us whether we want to recover tabs that were opened before the browser crashed. Curiously there is only one button for us to answer…
Transparent gifs. I hear you sigh as your mind wanders back in time, to days long gone, when there where only 2 browsers and both of them had layout quirks that demanded the use of transparent 1×1 pixels gifs. every. where.
The return of the GIF
GIF was huge in the early days of the web. It was the one format that could be used to send images over slow connections and still have an acceptable page load time. It could do animations! And most of all, as web developers, we used them to artificially inflate our table cells so IE and Netscape would leave them as we intended them.
Then, some company claimed ownership of the format and people called for the banning of GIFs from the internet. Browsers grew up and the 1×1 transparent GIF faded away…
Web Storage API
HTML5 gave us this beautiful thing: consistent cross-browser local storage of ‘vast amounts’ of data on the visitor’s computer in the form of the Web Storage API. The default quota is 5MB, which is huge compared to cookies. I’m using it in the development of Bridal App. It allows the app to respond near-instant to user actions and continue to function even when offline. All modern browsers support it (on desktop as well as on mobile) so life is great. Right?
For some years now I’ve had the feeling that Firefox’ development has been going in the wrong direction. The software just never seemed to get back to the speedy experience we had with Firefox 2.
Another project I have had this feeling with is Eclipse.
It turns out that both these projects share a commonality: long, standing, unresolved issues to core functionality.
Have a look at this Firefox issue:
Bug 195361 – Can’t select text from disabled form fields
Back in september of 2010 I wrote a blog post about the Breaking Float rendering bug I found in WebKit, the rendering engine backing the Safari, Chrome and Opera browsers. In april of 2013 the WebKit team landed a patch which would eventually end up in the browsers relying on WebKit. Case closed right? Well, not quitte…
Blink brought it back
Coincidentally, in the same month that the Breaking Float bug was fixed, Google forked WebKit into a new project called ‘Blink’. And the fix to the Breaking Float bug was not in the fork. Over time, the Chrome browser switched over to using Blink instead of WebKit, bringing back the issue in all it’s breaking glory.
Apache 2.4 running on Windows hangs when Internet Explorer 10/11 is used to acces pages on it.
I had a very reproducible scenario: Only the first request to Apache coming from IE would work, all other subsequent requests would hang, in all browsers. As long as I did not use Internet Explorer there did not seem to be a problem, but once that was used everything would hang until Apache was restarted. Strangely enough after I had applied the fix (see below) I reverted it to do some more testing but was not able to reproduce the problem anymore…
Ladies and Gentlemen! For tonight’s fight, in the blue corner, weighing in at two and a half billion users, Wooorldwiiide Weeeebbb BROWSERS!!!! And in the Red corner, weighing in at hundreds of corporations, institutions and individuals, Oooficiaaall Weeeebbb STANDARDS!!!
Browsers versus Standards may sound a bit weird, but if you are a web developer like me you may have had the feeling too sometimes that some standards as described by the organizations governing them actually seem to be hurting your attempts to write cross-browser compatible web pages. For example:
Phasing out basic elements such as
u and leaving the
iframe element out of the
Limiting the use of text/html for xhtml to xhtml 1.0 only.
Trying to replace the
img tag with the more generic
For some years, after finishing HTML 4, W3C has been, in my opinion, straying from ‘the path’ when they ventured into the dark woods of xhtml 1.1 and 2 and it’s many modules. They had made HTML so complex, just trying to figure out what doctype to use for your new HTML template could lead to hours and hours of internet research, leaving you more confused afterwards than you started out.