IE7 Beta 2 preview download

I knew it was due out soon,

So do I take it to mean that IE7 isn't going to be heavily restricted by that I mean is going to be truly stand alone, not just tied to Vista.

Does this beta overwrite existing install ? and it only runs on SP2 not all of us are mad enough to risk the chaos that ensues with this update pack.

I'll have to get a spare box updated to SP2 and install on that.

How does IE7 Beta seem to handle the height property as this still seems to be an area impossible to track down any hard facts on.

Hi Hugo,
It from MS, and it did overwrite IE6.
I am not sure how height or any other properties are handled, although it looks like the 'after' pseudo class is not supported. I downloaded the front page of the forum and associated files, to fix a minor problem, and IE7 was kind enough to rewrite my style sheet replacing 'after' with 'unknown'.

Well on the good side I have made a quick fix for the forums index page that hopefully wont mess up any other browsers.

On another forum a lot of people are uninstalling it. One guy said he got the BSOD. I can't install it because of some software I can't go without.

Wrong! FF 1.0.7 on OSX: the index page topic panels are now all a fixed height that combines to make a deeper page than previously i.e. they are not shrinking to suit the contents, and the blue keyline above the forum (sub-section) titles is missing in some sub-sections.

Hi Roy,
If only IE would support min-height :?
I had to apply fixed height values to get it looking reasonable.
Now that the Holly hack no longer works on IE7 I will have to find another solution.
Think of all the opportunities this provides young CSS guru's to find new hacks.

A quick search turned up this for IE7 Beta 1 - http://weblogs.asp.net/jgalloway/archive/2005/08/12/422335.aspx.
Not sure if the same still applies.

I though IE7 was supposed to support all the CSS2.1 properties. Granted height was still going to be expanding - but thats not so much of a problem if min-height works.

You're saying that all CSS2.1 properties doesn't include min-height (and presumably max-height, min-width and max-width). I wonder if there are any other CSS2.1 properties that somehow fall outside the "all". :roll:

By 'all', I think they were referring to css2 selectors, not properties. It's been a while since I could make myself look at the silly self-serving blog posts, so I may be mis-remembering. It was my impression that it came down to

• support all? css2 selectors
• fix the bugs listed at PIE
• add no support for any properties not already supported (this css thingy is a passing fad, right?)
• no support for svg
• no xhtml support (hasn't been around for long enough for anyone to really know how to handle it)

Inspiring, no?
cheers,

gary

[/]

I just did a little testing and can confirm the following selectors work:
Child: div > span
Adjacent: div + span
Attribute: img[title]

Support for * html has been removed so it can no longer be used to distinguish between IE and others.

The before and after pseudo elements still don't work.

margin-left:auto; and margin-right:auto; now center the element like they should.
Still need to use text-align:center for <= IE6.

Sounds like 6.1 or maybe 6.5 by another name. Same engine, same "features", bugs fixed, a few extra bells, no whistles.

I guess we should be grateful for the bug fixes.

There's a lot of complaining on SitePoint's forum about new (not necessarily rendering) bugs being introduced.

Browsercam have the Beta2 up and running in both the screen capture service and the remote access service if anyone wants to make use of it. Might be a good time to take out the free trial period...

Remote access doesn't work with the free trial does it?

nope but the screen capture does - should act as a gross error check.

Chris surely that's no solution to anything?

As I remember from that garbled confused blog, min/max-widths would be supported, but that they had no intention of changing the behavior of the height property, which is highly irksome of them.

If the pseudo elements :after and :before are not supported and the html star selector disabled then your clearfix technique is blown out of the water isn't it Tony and were left with either having to use conditional comments or overflow hidden (not always suitable)

As you say Chris this just ain't going to be the radical fixed browser that some believe Ok they are fixing a few of the daft bugs, slap on the back for fixing up errors in their coding -not, can't congratulate them on that score. They hinted that selector support would be vastly improved and include most of CSS2 but they qualified that with remarks about not introducing all of the newer stuff (CSS2.1 and up) as it was not formalized. What really is the problem, they have the Mozi example of fairly comprehensive selector support, what reason is there not to include stuff that is now pretty well established.

Why really are they intent on not including a full XML parser, there were some idiotic explanations on that along the line of how vastly difficult it was to implement and that they didn't want to do a bogged job, yet according to many the XML parser that Mozi uses is far easier to write than the tag soup version why do MS want to block XML in this fashion?

I've said it before and I'll say it again IE7 is going to do no more than introduce a whole heap of new problems to contend with, it just represents a new browser that we have to consider when coding .

Oh I forgot we should be jumping for joy as full support for png is introduced, 5 years to get around to adding png support, great.

I suppose that we should bear in mind that this is in beta, although surely it must be very close to final release, new bugs ok as long as they are willing to patch them with updates unlike past attitude.

I'm definitely not going to download this beta as I have absolutely no faith in MS ability to perform critical changes to their own OS without managing to screw things up, and mention of BSOD is enough to put me off.

Hugo.

Forgive me if I'm competely technically wrong on this but could it have to do with MS's development of future apps using XAML?

IE7 beta sounds like a load of crap.

FF 1.5 introduced SCG support, and it is fantastic. I would have thought that IE7 would do the same.

I also would have thought that IE7 would have support for as many, if not more, css selectors and properties as the Fox.

I, for one, won't be downloading it. Even if I decide tio upgrade to x64, or Vista when it's out, unless IE7 comes preinstalled, I won't use it.

Hugo,

My reading is that clearfix is not required for IE7. Its still implementing its expanding block feature - so it will automatically surround floats.

The holly hack is mainly used to get elements into hasLayout to avoid bugs that occur when an element doesn't have hasLayout. **IF** those hasLayout bugs have been fixed then IE7 "**shouldn't**" need the holly hack.

The problem of missing min-height and no "*" selector is we may no longer be able to get "min-height" to IE7 via a targetted "height" hack. That means no cross-browser footer at bottom page designs.

You can tell I'm an optimist can't you.

This all goes to show that M$ are continuing to try and sabotage the development of the web. If web applications forge ahead in popularity then M$ will lose sales of desktop applications, plus people will have less reason to use Windows over other systems. For as long as this scenario is just around the corner (it is), M$ will continue to try and make life difficult for web developers and designers. It's as clear as the window in my office!

On a brighter note, the message is very slowly starting to sink in on an ever growing scale. I am noticing that clients are increasingly more open to the idea of non-M$ software and many are now seriously considering a change of OS. The future is bright; the future is not blue!

Not blue!

It just may well be. Big blue is a huge supporter of Open Source projects

Chris: At present, IE6 does not enclose its float children unless hasLayout is true. Since an overwhelming number of the parents are dimensioned, we don't notice that it's required. Look at the examples in my enclosing float elements demo.

What will IE7's rendering be? ¿Quien sabe?

cheers,

gary

Yep Big Blue does indeed support open source, is it not them that are ploughing money into the Linux development community in an attempt to finally produce a Linux distro to fully rival Windows ease of use i.e something that can be installed in an hour rather than five and a weeks worth of tweaking (Gentoo Linux) better still though, as Apple are now porting the OSX to Intel processors, would be if they released it as a standalone OS then I 'd have copy like a shot.

Chris your right but Explorer only surrounds floats if the parent is in hasLayout mode, the holly hack was originally devised to feed height of any value to trigger that height expansion as it was thought that this was the sole trigger that height was required, this was before hasLayout had been discovered it was only after hasLayout had come to light that it was realised that in fact the holly hack actually was setting hasLayout =true.

If IE7 corrects hasLayout yet we need to enclose floats and IE7 does not work with :after then we are left with overflow:hidden or feeding height in conditional comments, saying that though will IE7 add support for the display table family ?

I admit to a certain amount of head spinning on this though and have probably missed an obvious point.

At the end of the day my retort to them is just fix the dam thing up! I'm entirely fed up to the back teeth with hackery and trickery to get CSS to work cross browser, bored witless with it.

HB this would seem an obvious conclusion that MS just want to stall full implementation of XML. When they offered their reasons for not implementing the parser a number of people, clearly fairly knowledgable, disagreed stating that in actual fact it was relatively simple to implement a XML parser, needless to say the IE team did not seem willing to argue and defend their position further.

roytheboy what have you been taking? "The future is bright; the future is not blue!" last time I looked it still looked fairly gloomy with many trials and tribulations to overcome before that happy place where everything works and MS has faded into a distant blurry memory.

You are right though there is a gradual dawning that MS is not the be all and end all, but we need that OS that works on as friendly a novice level as Windows yet with the power of a proper Unix heart for the nerds, Apple the future lies with you release that damned OS and watch the world change.

Hugo.

I remember saying that a while back but it didn't seem likely then. If Apple can actually release that OS to the world and implement their excellent marketing strategy along with it then we stand a chance.

So it should be - "the future's bright, the future's all white and shiny!"

Well there has to some reason behind porting the OS to the Intel platform. If Apple allow me to build my own box and install OSX on it I'll promise to buy one of their extremely cool and funky widescreen monitors to go with it.

When I mentioned 'blue' I meant the Microsoft blue

Apple have switched to intel chips because they need more power for less heat i.e. for their Powerbook range. However, Apple have been quite unpredictable of late, so who knows what they might do with their OS from here on in. I'm sure they'll see sense in the end [-o<

They didn't feel they were getting what they deserved from Motorola, especially in terms of product development (which is Roy's point). The rumblings started a few years back when Motorola was going nowhere. Motorola can probably afford to be more generous now.

Hi Hugo,

I'm not too keen on conditional comments so for now I will use IE's failure to correctly support !important.

height: auto !important;
height:1%; /*ie and browsers that don't support !important correctly */

That will get IE including IE7 beta 2 to contain the floats and other browsers can use :after;
I still need to test this out in mac browsers.

Well your forum listing is back to how it used to look - on FF_1.0.7_Mac and Safari_2.0.3_Mac

Tony there is another way to trigger hasLayout (apart from zoom) that IE7 will honour and it includes using the IE/Mac fix for the easy-clearing method.

While you're testing can you test if this still works on IE/mac..

.clearfix:after {
content: ".";
display: block;
height: 0;
clear: both;
visibility: hidden;
}

.clearfix {display: inline-block;} /* for IE/Mac and to trip layout */

/* Hides from IE-mac \*/
* html .clearfix {height: 1%;}
.clearfix {display: block;} /* reset to what it should be */
/* End hide from IE-mac */ 

I call it the 'tripswitch', as the display:inline-block; trips IE into triggering Layout then puts everything back to normal (except hasLayout).. so far it is supported in all IE, except IE5.0/Win but the holly hack can still be used for it if you want IE5/Win support

does it pass the mac test?

Coding hacks to work for IE5/5.5: $500
Coding hacks to work for IE6: $500
Recoding sites to work with IE7: $500
Coding once for Firefox and all other browsers: priceless

Just came across this:
http://www.brunildo.org/test/fenc7.html

Bruno has a bunch of grunt-work tests/demos. Good guy.

It comes to mind to ask what happens when the float's parent has layout, say a width, but no specified height. Does the parent enclose its child then, or not?

How does IE7 render my enclosing floats demo? Does it match modern browsers, IE6 or somewhere in between?

I have this gut feeling that MS may wish they hadn't even tried if the 'new' browser breaks all the IE6-centric pages out there.

cheers,

gary

Sadly it does not matter one bit if you use it or not. If it's rubbish and breaks stuff you'll still have to suffer the pain of working with it.

Unless that is you don't want to work as web dude I guess.

I for one will not be fixing sites if they break in IE7 - not in the early days at least. If we all rush to fix our designs then M$ will get away lightly and the rest of the world will be seen as being in the wrong. I know that we have no choice but to hack for IE6 or be berated by our clients, and the time will come when we will have to do the same for IE7, but I'm damned if I'm going to make life easier for M$ by minimising the trouble that their new browser is going to cause.

I WANT people to know that M$ have got it wrong. I WANT there to be a huge number of sites that are seen to break when it first comes out (if it's through fault of IE7). All M$ have to do is GET IT RIGHT and there wouldn't be any fuss. How many more millions of manhours across the planet are going to be wasted due to their incompetence or sabotage (whatever)

gary

Gary it's working just fine (looks same in both IE6 and IE7) this is because inline-block is triggering hasLayout to true for IE7 ~ I see you've added the display: block; into the clearfix rule.. this keeps Opera right (as it recognizes inline-block), hence the *reset* rule for the trip to work safely,

The ingredient were already built into EasyClearing by the IE/Mac fix.. the trigger could possibly be used for other situations too to avoid dimensions..
test demo: http://www.tanfa.co.uk/css/examples/haslayout/tripswitch-demo.htm

btw.. just for info.. did you know that if you explicitly set margins on the <p> elements (or whatever element) IE will magically start honouring the collapsing margin in the float, it only seems to wrongly collapse them if left to their own devices ~ adding this to your demo has the effect of making it look the same x-browser too though

Suzy?

.. just kidding about hiding, Yes it's me.. I don't get out as much as I should

??? Well! Isn't that special? That means the Holly Hack is not necessary for Tony's hack.

When you say "same in both", does that also apply to IE6's need for hasLayout because it fails on the display and overflow properties?

I did not know that. That will save a lot of frustration and will explain why I had varying results. If only IE would have just got it right to begin with …

"cssangel"? It would have been a Good Thing® to bring some of your rep and credibility with you, Suzy.

cheers,

gary

>>special
It might be, but I was hoping never to have to use it, however as always it's another tool that looks as if it might find a plce in the box! ~ I just thought I'd pop it in here to defend Tony's honour after reading that the EasyClearing method was trashed by the lack of the Holly Hack and * HTML hacks (on IE Blog IIRC), when in fact it's actually still working, albeit through serendipidity perhaps..

>>necessary
No indeed it might not be necessary UNLESS you want IE5.0/Win support, because IE5 does not support inline-block , you might still want to use the Holly Hack to backdate the height trigger for it..

put simply you now have a choice:
1. height is a hasLayout trigger for 5.0, 5.5 and 6, whereas
2. inline-block is a trigger for 5.5, 6 & 7..
3. or a mixture of both (as it is at present)

and remember that if you choose the inline-block method for other situations that the rules must be separate, see link example above. There would be no need to hide the second rule from IE/Mac, UNLESS you're actually using the inline-block part to fix something in IE/Mac, like the EasyClearing method does

Yes the two, IE6 and IE7, are succeeding and failing in exactly the same places, the widths you have in place for your IE6 tests are also working to trigger hasLayout=true for IE7.. but there is no improvement on overflow (which would save all this bother really )

Ah yes but where would the fun be actually it's a little known fact about the explicit margins methinks and although it fixes the float scenario.. it doesn't go the whole way to getting x-browser compliancy as far as collapsing margins are concerned. But having dug my head into the Collapsing margins recs. on more than one occasion I'm still not entirely sure who's right or wrong on that one yet..

Why thank you but as you can see from the join date, it was way back when.. before I earned any "rep".. and in those days I valued the anonymity having recent wounds from a forum (non-tech) where it was not advisable to be traceable. anyhow that was then and now I still believe in CSS and I still believe in Angels ~ so it's not that bad.. ~ and btw my name's not even really Suzy however I've adopted her,

hth, anyway
Suzy

btw.. If it's possible to change nicks here then I will if you want or perhaps I'll just put in my profile...

Should I have said something a bit more mysterious like:

Scarlett?

cssangel: Yeah, the collapsing margin thing will give you tired head. As I read it, IE creates problems where hasLayout and margins intersect. I shall have to do some testing to be sure of how to work the explicit margin with floats.

I'm pretty sure the theory is understandable. The practical is a little harder to suss out. On which element is what margin? Bruno Fassino has published quite a demo on IE/Win: margin collapsing and hasLayout, that you've probably seen, but is worth a look by anyone wanting to have familiarity with this vexing issue.

Didn't note the enrollment date, just the recent appearance and post count. :oops:

The aliases? No matter. Anyone who visits WWF will soon know the rep of Suzy, whoever she is. Here's hoping you hang here regularly; we can always use the expertise.

cheers,

gary

World Wrestling Federation?

Suzy definitely can flex some muscle, but no. It's the WebmasterWorld Forums SuzyUK is a css forum moderator, where she rules with an iron fist.

cheers,

gary

Sadly it does not matter one bit if you use it or not. If it's rubbish and breaks stuff you'll still have to suffer the pain of working with it.

Unless that is you don't want to work as web dude I guess. I'm going to disagree with this one. If a whole bunch of websites break using IE7 the obvious blame (even with the general public and hoi polloi) is going to be placed directly on Microsoft. The complaints are going to fly and Mr. Gates is going to be unhappy and fire a lot of people.

My point entirely. Do us all a favour and delay fixing your sites for IE7 as long as you can. Let the sh*t fly for a bit.

<rant>Like Hugo, I'm sick to death of hacking for IE. Sick to the back teeth. I've been leaving CSS alone for a while whilst burying my head in some pretty complex php scripting, and what a lovely break it's been. CSS ought to be easy and logical but IE makes it a complete pain. It's great that you're all getting to grips with IE7's failings (or corrections depending on your point of view), but what hope have our busy webmaster colleagues got of ever understanding it all? I've got a fairly good grasp of the intricacies of CSS but if I would rather deal with ultra-tricky, brain-aching enhancements to long php scripts than hack a site's CSS for IE then something is seriously wrong somewhere. My loathing for M$ grows every day! </rant>

but are krappy tag-soup table-alicious sites going to break in it? if not, then I wonder where that leaves us?

fending off our clients? working late trying to fix everything?

I like Roy's argument/reasoning...but we'll see...

interesting times ahead...

Sadly it does not matter one bit if you use it or not. If it's rubbish and breaks stuff you'll still have to suffer the pain of working with it.

Unless that is you don't want to work as web dude I guess. I'm going to disagree with this one. If a whole bunch of websites break using IE7 the obvious blame (even with the general public and hoi polloi) is going to be placed directly on Microsoft. The complaints are going to fly and Mr. Gates is going to be unhappy and fire a lot of people.

I did not so much mean fixing existing sites to be honest - I for one will be resisiting all pressure to fix stuff for as long as possible - I'll fail cause my boss will just shout at me and make me do it but hey ho - and I hope that it makes the average joe realise there ARE better things out there to be had for free.

I was more meaning that once it's "out there" and people are using it, which they will alas, even if it's totally bobbins, then we can't just say "I'm not using it, i'm not coding for it, let these new sites break in IE7".

fending off our clients? working late trying to fix everything?

I'm willing to bet that rubbish table-based sites won't break in it and I don't think our clients will be blaming MS - they'll be blaming us and be very reluctant to fork out any more money for fixes for what they'll see as our failure to build their sites "properly".

It's not right - but I think that's what'll happen.

A very good point - will it just be our CSS sites that break? Or will crappy nonvalid no-doctype font-ridden tagsoup pages break?

It's not right - but I think that's what'll happen.

that's what I'm thinking.

:sigh: