Google Chrome's Rendering

[Paul Komski]

Have been trying Chrome and it has many attractive features and great speed. However I have noticed that with these forums I get different rendering of some links - in particular those links that give me drop-down options (pics attached below). I thought it could be due to the absence of one line of source code.
<script type="text/javascript" src="clientscript/vbulletin_global.js?v=361"></script>
<script type="text/javascript" src="clientscript/vbulletin_menu.js?v=361"></script>The second of the two above lines (just above the title tag) is missing from the source code of Chrome but present in Firefox3, Opera9 and IE7 but may not be related since Opera has both lines of code yet the same link menu bar is akin to that using Chrome.

Do other's get similar results or maybe I am blocking something here.

[Paul Komski]

Just a little bump. This still perplexes me and it's only with chrome so far.

http://i37.tinypic.com/6jg4g9.jpg

Any ideas?

[mjc]

With Javascript disabled (ScriptBlocker) I get the same bar as Chrome.

[Paul Komski]

Interesting - and the same happens, I now see, in IE with Active Scripting disabled in the Security section of its Internet Options.

Safari has the same appearance as Chrome and what is bizarre (until I understand how javascripts are being parsed in Safari and Chrome) is that the line of javascript code that reads ( <script type="text/javascript" src="clientscript/vbulletin_menu.js?v=361"></script> ) just before the body tag is absent in Safari and Chrome's source code but is still present in IE with scripting disabled.

Perhaps it's to do with the V8 javascript engine that Chrome uses (http://www.codeproject.com/KB/library/Using_V8_Javascript_VM.aspx) but what I don't understand at all is why one line of the code is stripped from the source and why the other, nearly identical one, is left intact!

[mjc]

I gave up trying to figure out why Chrome does what it does about two hours after it was released...there are a couple of other things it does strangely, too. I can't quite remember what they are at the moment...

[Paul Komski]

The more I read about the javascript engine in both Safari and Chrome the less I think that either is involved in modifying the source code. The answer to the "problem" must lie in the way the server's php parses the code for different browsers (user agents if you prefer) requesting web pages.

Both SquirrelFish (http://webkit.org/blog/189/announcing-SquirrelFish/) when used in Safari and the V8 engine in Chrome seem to introduce huge speed boosts to both particularly for javascript-heavy web pages (such as GMail). It is obviously seen that such web programming will become both more intense and more commonplace to give much more dynamism to the web experience.

My own OP should now probably be rephrased to ask the new question of why the vBulletin code supplies different code to Safari and Chrome.

I must say I'm very impressed by Chrome's speed, its "clean" appearance and some other functions but must also take a look at a recent up-to-date version of Safari.

[Suzanne]

i hear that google chrome has a lot of deficiencies like the errors of discussing in this thread.. but i believe that google administration will figure it out soon and fix the errors.

[Paul Komski]

Hi Suzanne and Welcome to the PCGuide.

It's true that Chrome is still in beta so one cant be too critical. The same is not true of Safari, which has the same result and my own current belief is that what I thought was a browser problem is actually a php server problem. Without being able to read the php source it will remain unclear.

[mjc]

I haven't played with Safari...does it allow you to change the UA string?

If it does, I wonder what it would display like if you caused Safari to 'lie'...

[Paul Komski]

does it allow you to change the UA string?

If it does I cant find it. Having posed the question I then wondered if I could get Opera or Firefox to do the reverse and spoof as Safari (or indeed Chrome). Opera only has options for Firefox and IE and all I could find for Firefox was the UA Switcher Extension (http://chrispederick.com/work/user-agent-switcher/).