Issue with IE 5.5 under new board?

Quirk? So in other words the small business computing forums which are suppose to be using the same templates and code as here do not cause this quirk to surface but here the quirk does surface?

The two source captures do show somewhat different code being generated for the same thing in ie 6 vs ie 5.5:

As additional info...

If I open a reply in an IE 6 browser, save the code that was generated to produce the screen, then go open that (saved code) in a IE5.5 browser it displays properly/normally. On the other hand, if I open a reply (same one) in a IE 5.5 browser, save the code that was generated to produce the screen, then go open that (saved code) in a IE6 browser it does not display properly.

So something in the different code being generated when the browser is 5.5 is causing the problem (regardless of what browser version then opens that code).

The templates are never quite identical. I didn't find this issue on the vbulletin forum, but I did find other IE 5.5 issues and their response to the poster was to download 6. I'll post the question there and see what they suggest. I know they do test for version in the PHP code.

See what they say, because it is clear (and can be seen by doing a line by line comparison) that the code generated in/for the 5.5 browser is different than that for the 6.x. And as I said above, a test shows that if the code being generated for 6.x was used for 5.5 it would work properly (and the 5.5 code run in a 6.x browser behaves badly just like in the 5.5 browser).

If this is to indeed be identified as a "quirk" or such then we should be able to say what lines/commands in the code are the ones that 5.5 is not reponding to the same (or properly). Otherwise it is just as reasonable to assume it is a coding flaw.

They test for IE5.5 and serve it different HTML, precisely because of it's quirks. It was the single most flawed version of IE they produced. Even simple things like the CSS background-color property cause it to choke. However they haven't responded to me as to what else might break if I comment out the browser sniffer in the code. I know where the test for version is being done, I just don't know the repercussion of removing it.

Well it will be interesting to see what they say. It is certainly odd that they would test for 5.5 and do something different (vs 6.x) accordingly here when the 6.x code generated for the screen seems to work fine by all accounts when put in to a 5.5 browser. And it is also interesting that their (generated) code for 5.5 at the small business computing forums works fine as is. I can understand the potential need or desire to perhaps serve up different code for different browsers. But it just seems like the possibility that there is something wrong in what is being served up (as opposed to what it is being viewed in) might not be being given due consideration.

If in the end there is a decision that only certain browsers (including versions) will be supported and will enable members to use this site fully then such minimum requirements should rightfully be stated.

I don't think it's the browser sniffer per se, but some otherwise insignificant difference in the code on this page as opposed to the other site of ours. But my point to you is it doesn't need to be incorrect coding for 5.5 to choke on it. It may be entirely correct and 5.5 can still choke. That's what makes it nearly impossible to track down. It's not wrong, just wrong for IE5.5.

Understood. It may well be that it is something in IE 5.5 (though there is reason to question this). However, before this can be concluded the coders should be able to determine exactly what code it is that 5.5 is misbehaving on. Furthermore, it would just seem somewhat logical that since they are doing something different/special for 5.5 anyway, what they are doing now clearly doesn't work, and that there is (as can be demonstrated) something they could do to make it work, that requesting they make an appropriate change (even if it is to compensate for "a quirk") is not a huge stretch. In other words, if they know there is a quirk, they compensate for it. They should also consider the fact that if the code they are currently generating for 5.5 for the Reply page does not work right, yet the code they generate for a 6.x browser does work properly even in a 5.5 browser, then it would make sense for them to just produce common code for both browsers.

Well they must've started testing for a reason. There must be some known 5.5 issue they're trying to get around, I just don't know what it is, and they don't seem all that anxious to tell me.

Quote:

---

and they don't seem all that anxious to tell me

---

Oh boy! :D That's often a pretty good indicator that a) they don't know, or b) they don't want to deal with it. ;) So we'll have to continue to see what happens.

Well I bumped the thread once and they were a little perturbed at that.

I appreciate you efforts JP. Keep us posted.

Any update on this issue???

Bump...