Slashdot Mirror

...that doesn't impress me much.

As a web developer, this doesn't really do anything for me because:
1. Virtual PC was already free
2. An XP license is a negligable cost (if you don't already have an MSDN subscription)
3. This does nothing for IE 5 & 5.5

Maybe IE 5 & 5.5 are so long ago inside the Microsoft campus that they can be forgotten, however in the real world people still use them.

In addition Virtual PC is a headache because:
1. You need to boot the machine up which takes a while
2. You can't hit a local IP address
3. It is noticably slower than running natively (Core duo2 6400, 2Gb RAM)

When you compare this with how say firefox works, the Microsoft solution just doesn't compare well. With firefox I can install multiple versions and run them side by side.

In addition, this still doesn't answer why Microsoft chose to sic the lawyers onto the much more useful solution, which is closer to the Firefox style.

Full credit to Microsoft for the attempt, but it is a pretty half assed one at that.

Late response, but can I add that one of the best options around until recently was a hacked workaround to install all the versions of IE onto one box (here). Sure it was less than perfect (For example I had issues with cookies, making it impossible to test full applications but possible to test single pages). However it seems that Microsoft has decided to make life even more difficult for web developers by shutting this down. WTF?

This is a serious issue for web developers. We need to be able to run multiple versions of IE to test sites because web users still use earlier versions of IE. This is a particular issue because the different versions of IE do some things wildly differently (eg IE5 and 'undefined', a string as opposed to undefined which everyone else uses).

Sure I can run this under a virtual machine and kudos to Microsoft for making Virtual PC available for free. I shouldn't have to though. I can run multiple versions of Opera. I can run multiple versions of Gecko. I can run multiple versions of KHTML. I can't run multiple versions of mshtml.

Late response, but can I add that one of the best options around until recently was a hacked workaround to install all the versions of IE onto one box (here). Sure it was less than perfect (For example I had issues with cookies, making it impossible to test full applications but possible to test single pages). However it seems that Microsoft has decided to make life even more difficult for web developers by shutting this down. WTF?

This is a serious issue for web developers. We need to be able to run multiple versions of IE to test sites because web users still use earlier versions of IE. This is a particular issue because the different versions of IE do some things wildly differently (eg IE5 and 'undefined', a string as opposed to undefined which everyone else uses).

Sure I can run this under a virtual machine and kudos to Microsoft for making Virtual PC available for free. I shouldn't have to though. I can run multiple versions of Opera. I can run multiple versions of Gecko. I can run multiple versions of KHTML. I can't run multiple versions of mshtml.

George W. Bush is George H. Bush's son.
George H. Bush attacked Iraq and was a target of the Iraqis in that war.
Therefore, George W. Bush will attack Iraq.

See the difference? This latter example *behaves* logically -- it has 2 true premises, and a true conclusion. The conclusion would seem to follow, if we assume that W. wants to retaliate for "trying to kill my daddy" (which is an easy conclusion to jump to, and may even be true).

BUT, there is no reason to believe, on the basis of heredity alone, that George W. Bush will attack Iraq -- as this latter example suggests. This is "soft" logic. The conclusion requires a hidden assumption -- this is the essence of soft logic: one or more hidden assumption to reach a conclusion. Hard logic lays all factors out on the table (the factors may be way-off or wrongly-conceived, but they still must come together to form a specific, singular, no-arguments-about-it result).

Hard logic requires the use of control statements or math. Soft logic requires accurate premises and conclusions, but the premises needn't actually be directly-connected.

HOWEVER: seemingly-soft logic can actually be "hard" logic. For example:

The elevator's suspension cable will break if another person gets on.
Another person gets on the elevator.
Therefore, the elevator's suspension cable breaks.

If the first premise is true, and the second is true, then the third *must* be true. This looks like soft logic because it is stated in english and not in a programming language or math scribbles, but because it contains quantifiable elements and a strictly-defined path towards the conclusion which either must or must not occur, it is hard logic.

Likewise, hard logic can defy reason. Take numerology as the classic example: it's mathematically possible to "prove" that girls are evil. We know girls aren't evil, yet the math undeniably suggests it's true. The factors are all there, and structurally and computationally, the algebra does correctly conclude, on the basis of the parameters, that girls are evil.

Yet, if we look at this example of "hard" logic and dig beyond its superficial calculation procedure -- which is hard logic -- and examine the individual factors, we find that this is actually soft logic. Why? Because one of the premises is based on a quote which is actually just the opinion of various philosophers. Money isn't really "the root of all evil" (even though some emotionally believe so). Not, that is, unless you consider yourself a communist, because after all, money represents goods and services -- tangible things which, when viewed directly (and not through the pointer of money), we don't consider evil.

This last example is just the sort of "logic" that is employed by social scientists everywhere, particularly economists. Their process flows and math and statistical calculations may all be correct -- but the data they use has some bias or some flawed premise (whether intentional or not), causing the the results from these data to be skewed. That's why 2 economists c

There is no Windows box that lets you run IE5 and IE6 side by side, and this is actually a rather practical thing to do if you're a developer.

Are you sure?

Or instead, you can use this guy's script to apply the filters for you on IE. Include it once and all the PNGs on the page work properly. (Why MS couldn't do this themselves is really puzzling).

Well, apparently the Shuffle is no good as a food product.

You know that you can run multiple versions of IE on the same windows box? See this link.

Alpha chennel PNGs in IE 5.5+. Sure, it's a bit of a hack, but once it's in, you can pretty much code your site without worrying about it, AND it validates as XHTML 1.1 :^)

Too bad you can't do the same with IE. Oh, wait.

Also, as the previous poster said, once you've installed the VMWare Tools, you don't need to press and release Ctrl+Alt in order to change the VMWare window focus; you can seemlessly move your mouse between the guest and host OS's (among other functionality).

Or you could just use Sleight. Much easier to implement.

I don't think there are any decent Mac emulators around. There are, however, decent PC emulators on the Mac.

If that's not an option, then you can't really do anything about Mac/IE, as the Mac and Windows Internet Explorers use completely different rendering engines.

Safari is based around the KHTML engine, and so you can be fairly safe with that browser as long as you test in Konqueror.

Things like Browsercam aren't very helpful, as you can't interact with them, and a lot of bugs only show up when interaction takes place. But if you have no other option, like Mac/IE without owning a Mac, then it's better than nothing.

Even if you aren't bothered about other platforms, virtual machines like VMWare are useful. You can set up a range of them with different screen resolutions, font size settings, Javascript on and off, and so on, so you don't have to keep fiddling with your settings.

If you take the "fiddle with your settings" approach to testing, set up a second account on your workstation for just this purpose. That way, any plugins, settings, etc, that you use for normal day-to-day surfing won't interfere with your testing. Make sure you keep a checklist where you can tick off each combination of settings that you have tested against - you will miss combinations otherwise. You will probably find it useful to install multiple versions of Internet Explorer on the same machine.

Obviously, run your code through HTML and CSS validators, and possibly linters as well. It's a good idea to incorporate validation into your publishing routine - nothing invalid ever reaches the server. If you can't do that, it's a good idea to set up a validator to automatically spider your websites on a regular basis and report any errors to you via email. Alternatively, check out Ben Hammersely's validation RSS feed.

Works great for me. I have seen a few reports of some odd CSS issues that don't actually reflect what the browser would do if it were the sole browser on the system but everything I've compared against VirtualPC or the few times I've been able to compare a site against a native install has been accurate.

Most of PNG works in Internet Explorer (just the alpha channel is broken, and there are limited workarounds)

It is "kludgy", but not too bad. Check out this page for how to easily get alpha transparancies in your pages. Check out Dredg's Online Store for a real world example.

You CAN do the transparent png stuff with IE, you just neet to add a simple js file to your html file. Check out this page:
http://www.skyzyx.com/scripts/sleight.php

Even though it is an impressive list, you can still only test for the 5% of the webbrowsing population.
The rest use IE 6, 5.5 & 5 (an yes, they all have their own quirks which is, quite frankly, a massive pain in the backside)

On windows, there is now a way (http://www.skyzyx.com/downloads/) to have all of those on one machine. In addition, the gecko rendering engine is there. The only one missing is khtml, so if that'd be ported you'd make quite a lot of webdevelopers happy.

PS. On Linux, the same applies as Mac, but you might get wine to work with the stand-alone IE's.