Got equipped with SCREW ATTACK!
-
@ASheridan said:
I'm sure if they wanted to they could introduce updates to fix these problems, but they don't seem to care.
Why would they? IE7 is 6 years old. Or do you also think Firefox should go back introduce drop shadows in Firefox 2.0?
You seem to have trouble with basic software development concepts here. New features go into new products.
-
@ASheridan said:
@Sutherlands said:
Yes, actually, that code. The code where you add things like var contentDiv = $('#description')'.last().parent().create("div");Geez I'd hate to see your code.
Which code is that? The code where I'm not filling my HTML with empty div tags for each Javascript plugin I'm using? I hardly see a huge problem with a single extra line of Javascript just for one div.
That's much less maintainable than adding an empty div into html.
And how many plugins are you using that require their own separate section?
-
@ASheridan said:
And somehow I doubt that Microsoft not making updates for its browser versions has anything to do with its concern for businesses that are afraid to update.
What are you talking about? IE7 has updates available, one is called IE8, another is called IE9, and yet another is called IE10. That one's fresh and new.
I really honestly don't understand what you're getting at here, especially since bjolling spelled out the issue for you. Do you honestly believe that if Microsoft made a new version of IE7 that fixed the drop shadows, call it IE 7.5, do you honestly believe a company that won't allow IE8 in their update system will allow IE 7.5 in it? Why do you believe that? Do you have any evidence to back up that belief?
Look, the problem you and bjolling are dealing with are rock-stupid IT departments. bjolling is smart enough to realize that is not Microsoft's fault, nor is there anything Microsoft or anybody else can do about it. (If these companies were using Linux, they'd be on Ubuntu 2008. If they were using OS X, they'd be stuck on 10.3. Windows is a non-issue here.)
And the thing that really bugs me is that none of these are functional changes in how the browser works. You're talking about drop shadows and PNGs. Graphical bullshit. You can write the exact same application no problem, it just doesn't look 100% identical without some extra work. Boo-hoo.
HTML5 is the only thing that's brought actual functional changes to how browsers work (and even then it's 99.9% backwards-compatible), since then the last change was XmlHttpRequest in IE 5.0. Your environment's plenty stable, web developers are just crybabies about it.
-
@blakeyrat said:
Why would they? IE7 is 6 years old. Or do you also think Firefox should go back introduce drop shadows in Firefox 2.0?
Why would they release updates for IE9? Simple, because the browser isn't that old. I wasn't really talking about how they should release updates NOW for the very old browsers. Of course, you weren't really reading what I'd written properly and were listening to those little aliens on your should whispering in your ear, so you decided to make up an answer to a question I didn't ask. Well done, you're an idiot.You seem to have trouble with basic software development concepts here. New features go into new products.
-
@blakeyrat said:
What are you talking about? IE7 has updates available, one is called IE8, another is called IE9, and yet another is called IE10. That one's fresh and new.
They're new major versions of the IE browser, not updates to versions. You seem to have trouble with basic software development concepts here.@blakeyrat said:
Do you honestly believe that if Microsoft made a new version of IE7 that fixed the drop shadows, call it IE 7.5, do you honestly believe a company that won't allow IE8 in their update system will allow IE 7.5 in it? Why do you believe that? Do you have any evidence to back up that belief?
Do you honestly believe I'm only talking about IE7 here? Why do you believe that?
@blakeyrat said:
Look, the problem you and bjolling are dealing with are rock-stupid IT departments. bjolling is smart enough to realize that is not Microsoft's fault, nor is there anything Microsoft or anybody else can do about it. (If these companies were using Linux, they'd be on Ubuntu 2008. If they were using OS X, they'd be stuck on 10.3. Windows is a non-issue here.)
I think you're confused here. I'm not blaming the fact that companies don't allow software updates on Microsoft. I never fucking have. I am saying though, that it would benefit some if Microsoft released updates to their browser that fixed bugs in their browser. Read what I wrote you fucking idiot, and stop listening to your little shoulder aliens.@blakeyrat said:
And the thing that really bugs me is that none of these are functional changes in how the browser works. You're talking about drop shadows and PNGs. Graphical bullshit.
I mentioned two things that first came to my head. Sorry, did you want a comprehensive list of everything that's wrong with IE? Don't be so fucking stupid.@blakeyrat said:
You can write the exact same application no problem, it just doesn't look 100% identical without some extra work. Boo-hoo.
Yes, you can write the exact same application. What takes the time is then making the necessary changes for every version of IE that absolutely has to be supported, and working around the plethora of bugs that exist in every bloody version of IE. It would be much nicer if each version of IE had the same bugs, but each version breaks in little tiny ways that are a pain in the fucking ass.@blakeyrat said:
HTML5 is the only thing that's brought actual functional changes to how browsers work (and even then it's 99.9% backwards-compatible), since then the last change was XmlHttpRequest in IE 5.0. Your environment's plenty stable, web developers are just crybabies about it.
Yeah, because Javascript is not functional at all, and doesn't have weird bugs in every version of IE.
-
@Sutherlands said:
I tend to keep my Javascript separate from my HTML, so no, my HTML wouldn't have stuff like that in. In my comment I was referring to the HTML which I'm trying to keep clean and semantic as possible. Sure I guess the odd empty element isn't the worse thing in the world, but if it's avoidable and won't cause a headache, what's wrong with avoiding it? Your example of that code being less maintainable if the document structure changes; how often does the DOM change so dramatically without anyone working on the Javascript for it too?Yes, actually, that code. The code where you add things like var contentDiv = $('#description')'.last().parent().create("div");
-
I'm working for a company that won't let its users upgrade beyond IE8. Why? Because there are critical infrastructure web applications (each of which cost a small fortune) that were not built with web standards. Somehow these applications fail to work even under compatibility mode in newer IE versions. This is a WTF, but not the main one.
The reason why users are not allowed to upgrade past IE8 is that IE9 and IE10 refuse to install alongside IE8 (in any supported fashion). IE, to my knowledge, is the only browser with this major and crippling limitation.
I could stop pulling my hair out about polyfills and browser hacks, and move our public-facing website into the 21st century if IE would just allow our users to run an older version for specifically for our internal legacy web applications.
-
I thought IE10 was exclusive to Windows 8, so you'd have to do some serious hacking to even think about installing it on a system that had IE8...
-
Yet another reason we can't install them side-by-side!
Seriously, though, that's the actual real reason we won't be rolling out Windows 8 here: it doesn't run IE8.
-
@ekolis said:
I thought IE10 was exclusive to Windows 8, so you'd have to do some serious hacking to even think about installing it on a system that had IE8...
They're supposed to release it for Windows 7 this month, so when that happens, you can (because Windows 7 originally shipped with IE8).
I, however, am going to keep using IE8 on my home computer until Microsoft stops forcing ClearType, or I decide I don't want my eyes anymore (because that's what will end up happening if I have to look at that mess of colored pixels).
-
Microsoft and the W3C have always had a strained and complicated relationship. Frankly I think Microsoft holds the high ground, since when you look at the histories of it, it's exactly as blakey says, Microsoft is always first to the party with a given feature, which is then implemented in the spec differently, rendering IE the non-compliant browser. I used to be on the hate-IE bandwagon (and pre-IE9, there were many legitimate reasons to hate it) and am still an avid Chrome user, but nevertheless, Microsoft isn't golden. I've never understood why they only update their browser once in a blue moon, maybe once or twice between OS versions.
And even worse is the new trend where each new version of Windows has a new version of IE, which isn't backported at all. If Microsoft goes that route they're just going to drive more people off of IE, and considering they just slipped behind Chrome recently, getting more consumers off of IE isn't a good idea. There is not a single reason IE10 couldn't be used on Windows 7, or IE9 couldn't be used on XP, the only explanation is good old fashioned greed, similar to how they tried to require Vista to play new PC games back when that first came out, which turned into a PR nightmare.
Frankly, whenever you start to take a real granular look at most of these little "wars," both parties involved look stupid.
-
@Master Chief said:
Microsoft and the W3C have always had a strained and complicated relationship. Frankly I think Microsoft holds the high ground, since when you look at the histories of it, it's exactly as blakey says, Microsoft is always first to the party with a given feature, which is then implemented in the spec differently, rendering IE the non-compliant browser. I used to be on the hate-IE bandwagon (and pre-IE9, there were many legitimate reasons to hate it) and am still an avid Chrome user, but nevertheless, Microsoft isn't golden. I've never understood why they only update their browser once in a blue moon, maybe once or twice between OS versions.
And even worse is the new trend where each new version of Windows has a new version of IE, which isn't backported at all. If Microsoft goes that route they're just going to drive more people off of IE, and considering they just slipped behind Chrome recently, getting more consumers off of IE isn't a good idea. There is not a single reason IE10 couldn't be used on Windows 7, or IE9 couldn't be used on XP, the only explanation is good old fashioned greed, similar to how they tried to require Vista to play new PC games back when that first came out, which turned into a PR nightmare.
Frankly, whenever you start to take a real granular look at most of these little "wars," both parties involved look stupid.
How exactly is Microsoft the first browser company to implement border-radius? Or border-image? Or opacity? Sure, they had their own proprietary version of the last one, if you like typing out progid:DXImageTransform.Microsoft.Alpha(Opacity=50)
-
@Ben L. said:
@Master Chief said:
Microsoft and the W3C have always had a strained and complicated relationship. Frankly I think Microsoft holds the high ground, since when you look at the histories of it, it's exactly as blakey says, Microsoft is always first to the party with a given feature, which is then implemented in the spec differently, rendering IE the non-compliant browser.
How exactly is Microsoft the first browser company to implement border-radius? Or border-image? Or opacity? Sure, they had their own proprietary version of the last one, if you like typing out progid:DXImageTransform.Microsoft.Alpha(Opacity=50)That was Microsoft's position. They lost it to Mozilla when they decided that the HTML suite was "done", and Mozilla lost it to Google because Google actually develops web apps and decided to be the change they wanted to see™ instead of just trying to force Gears onto everybody.
Per your question, border-radius was Mozilla (as the propritary -moz-border-radius), border-image was Mozilla (as the propritary -moz-border-image), and opacity was Mozilla again (shockingly as just opacity instead of -moz-opacity).
Interesting side note here, while mainline Webkit doesn't use -webkit-border-image anymore, mobile Webkit does.
You guys want to rant about a browser that doesn't implement a lot of web specs and that people don't upgrade (because it's tied to the Android version)? Rant about Mobile Safari.
-
@Master Chief said:
Microsoft and the W3C have always had a strained and complicated relationship. Frankly I think Microsoft holds the high ground, since when you look at the histories of it, it's exactly as blakey says, Microsoft is always first to the party with a given feature, which is then implemented in the spec differently, rendering IE the non-compliant browser
I think this at least in part could be the way that it was originally implemented wasn't the best going forward. I read the stuff going back and forth on the WhatWG mailing list for HTML5 and its brethren, and sometimes there's a lot of discussion from a lot of parties (browser vendors included) so that the spec that comes out is better for everyone concerned. Sure they don't get it right all the time, but it would at least explain why IE's new features are re-specced by the W3C. I mean, consider their DirectX CSS filters, that most often look awful, don't always play well with each other, and have a syntax that sometimes makes an assembly programmer want to weep.
-
@MiffTheFox said:
The follow a slightly different release cycle, the same way that Firefox & Chrome for mobile and desktop do.Interesting side note here, while mainline Webkit doesn't use -webkit-border-image anymore, mobile Webkit does.
You guys want to rant about a browser that doesn't implement a lot of web specs and that people don't upgrade (because it's tied to the Android version)? Rant about Mobile Safari.
-
@MiffTheFox said:
You guys want to rant about a browser that doesn't implement a lot of web specs and that people don't upgrade (because it's tied to the Android version)? Rant about Mobile Safari
I hear they have a really sharp javascript debugger.
-
@ASheridan said:
@MiffTheFox said:
The follow a slightly different release cycle, the same way that Firefox & Chrome for mobile and desktop do.Interesting side note here, while mainline Webkit doesn't use -webkit-border-image anymore, mobile Webkit does.
You guys want to rant about a browser that doesn't implement a lot of web specs and that people don't upgrade (because it's tied to the Android version)? Rant about Mobile Safari.
I literally cannot upgrade the web browser on my phone without hacking it, as Google's dropped all support for my device (Nexus 1) to focus on their ~shiny new Samsungs~. The fact that XDA Developers got Android 4.x working almost just fine (except the graphics were bugged up last time I tried to upgrade, which is becuase Quallcomm or whoever was responsible for the SOC doesn't want to release drivers) says a lot for Google's piles of WTF.
Instead I get dumped with a shitty old version of Mobile Safari that adds three copies of pages to the history on certain sites like Tv Tropes, doesn't ever pick up on title tags (so history and open windows show up as "http://en.wikipedia.org/wiki/... - http://en.wikipedia.org/wiki/..."), and once crashed my phone hard trying to render the TF2 wiki (taking my SD card with it).
Part of why I resolved not to get another "smart" device ever. (Unless it's something like the Thinkpad Tablet 2 or the Surface Pro where core applications aren't integrated into the OS.)
-
-
@MiffTheFox said:
Google's dropped all support for my device (Nexus 1) to focus on their ~shiny new Samsungs~.
They’re really, really shiny, though. I just got a Note II five days ago. While I’m not sure how I’m going to carry it when I’m not wearing carpenter jeans, it’s a nifty piece of kit. The s-pen makes it actually useful for quick note taking.
-
@MiffTheFox said:
I literally cannot upgrade the web browser on my phone without hacking it, as Google's dropped all support for my device (Nexus 1) to focus on their ~shiny new Samsungs~.
Have you tried a different web browser? I had an HTC Desire which is pretty dated now, and Opera (Mini & Mobile), Firefox & Dolphin (the best of them in my opinion) are all installed on it.@MiffTheFox said:
Instead I get dumped with a shitty old version of Mobile Safari
Not surehow you've got Safari on an Android phone. The stock browser uses webkit, but that doesn't make it Safari, it's just the rendering engine.
-
@ASheridan said:
@MiffTheFox said:
I literally cannot upgrade the web browser on my phone without hacking it, as Google's dropped all support for my device (Nexus 1) to focus on their ~shiny new Samsungs~.
Have you tried a different web browser? I had an HTC Desire which is pretty dated now, and Opera (Mini & Mobile), Firefox & Dolphin (the best of them in my opinion) are all installed on it.@MiffTheFox said:
Instead I get dumped with a shitty old version of Mobile Safari
Not surehow you've got Safari on an Android phone. The stock browser uses webkit, but that doesn't make it Safari, it's just the rendering engine.Interestingly, the default Android browser identifies itself as Safari.
-
@Pidgeot said:
or I decide I don't want my eyes anymore (because that's what will end up happening if I have to look at that mess of colored pixels)
Tweak it. I don't have coloured pixels. Tweak it.
-
People who mention ClearType are trolling.
-
@pkmnfrk said:
Interestingly, the default Android browser identifies itself as Safari.
They also identify as a mix of KHTML and Mozilla, despite webkit no longer being KHTML and not being Mozilla, but if you check this list, they do all identify as Android. http://www.gtrifonov.com/2011/04/15/google-android-user-agent-strings-2/
-
@ASheridan said:
@pkmnfrk said:
Interestingly, the default Android browser identifies itself as Safari.
They also identify as a mix of KHTML and Mozilla, despite webkit no longer being KHTML and not being Mozilla, but if you check this list, they do all identify as Android. http://www.gtrifonov.com/2011/04/15/google-android-user-agent-strings-2/Well that's because it's masquerading as Netscape so that user agent sniffers won't treat it as Internet Explorer and send it "dumbed down" content it can actually handle.
-
-
@MiffTheFox said:
@ASheridan said:
@pkmnfrk said:
Interestingly, the default Android browser identifies itself as Safari.
They also identify as a mix of KHTML and Mozilla, despite webkit no longer being KHTML and not being Mozilla, but if you check this list, they do all identify as Android. http://www.gtrifonov.com/2011/04/15/google-android-user-agent-strings-2/Well that's because it's masquerading as Netscape so that user agent sniffers won't treat it as Internet Explorer and send it "dumbed down" content it can actually handle.
Wow, 1998 is calling and wants its technology back.
-
@MiffTheFox said:
Probably very likely then why the user agent for Android contains "mobile Safari", because crappy developers are writing crappy "mobile" websites, and think that mobile is synonymous with iOS.Well that's because it's masquerading as Netscape so that user agent sniffers won't treat it as Internet Explorer and send it "dumbed down" content it can actually handle.
-
@pkmnfrk said:
Wow, 1998 is calling and wants its technology back.
You get the time machine ready, I'll start gathering up as many Mozilla/5.0's as I can find.
-
@dhromed said:
@Pidgeot said:
or I decide I don't want my eyes anymore (because that's what will end up happening if I have to look at that mess of colored pixels)
Tweak it. I don't have coloured pixels. Tweak it.
I have tried tweaking it. Repeatedly. On multiple different monitors (all of them LCD). Every single time, the result has been literally pain-inducing (and yes, I really do mean literally).
Also, if you don't have coloured pixels, you don't have ClearType, because the entire point of ClearType is using subpixels to increase the horisontal resolution, which implies that some pixels will have to be a different colour than either the background or the text colour. You may not be able to see those pixels, but a few of us (including me) do (yes, at normal viewing distances).
If you only have shades of grey, then it's anti-aliasing, not ClearType (and while anti-aliasing is *way* better than ClearType, it still looks awful compared to full-pixel rendering).
-
@Pidgeot said:
Also, if you don't have coloured pixels, you don't have ClearType
I think I kind of know what Cleartype is. :)
@Pidgeot said:and while anti-aliasing is *way* better than ClearType
AA is the general technique. Cleartype is one method of AA. Get your terms straight.
Cleartype's most obvious shortcoming is the lack of a vertical axis which means that slightly sloped horizontal lines look like pixelated shit. :\
This was supposedly fixed in Win7, but only for applications that use WPF or something, and I don't know what's what in Win8.
@Pidgeot said:
full-pixel rendering
That's called aliasing.
@Pidgeot said:
You may not be able to see those pixels, but a few of us (including me) do (yes, at normal viewing distances).
That's unusual, but okay. Aliased text for you it is!You know what else is funny? As I've fulminated over and over again here, I have my screens in portrait mode, so the LCD subpixels are turned as well, but nonetheless cleartype still works! (though I tweak it differently from landscape orientation, of course)
-
So tired of the endless Cleartype/AA/etc. arguments. Can we just go back to fucking vector monitors already?
-
@Weng said:
So tired of the endless Cleartype/AA/etc. arguments. Can we just go to fucking vector monitors already?
My monitor has 10 pixels. 5 by 2 widescreen.
-
@Weng said:
Can we just go back to fucking vector monitors already?
I'm not sure this is an improvement:
-
@Weng said:
So tired of the endless Cleartype/AA/etc. arguments. Can we just go back to fucking vector monitors already?
Because everyone loves bulky CRTs!
-
@bjolling said:
with Internet Explorer where it's very important to the business people that the browser is stable
I started strongly doubting the factual accuracy of this post about here.
-
@MiffTheFox said:
@Weng said:
Sorry, you failed your Detect Sarcasm roll.So tired of the endless Cleartype/AA/etc. arguments. Can we just go back to fucking vector monitors already?
Because everyone loves bulky CRTs!
-
He doesn't mean stable in the sense of not crashing. He means stable in the sense of not getting a new release every two weeks.
-
Just found out about this wonderful little IE
8 and belowbug:http://www.thecssdiv.co.uk/2009/08/another-weird-ie6-bug/
Basically, if you like to minify a bunch of CSS files into one to reduce the number of requests a browser has to make to your server, think again if you happen to use a lot of selectors (which is likely in even a moderately complicated design) then you might hit the limit that IE
8/7/6has with CSS selectors.I mean seriously? Where does this limit even come from? It's not even like the number 4096 makes sense. And Microsoft didn't happen to get round to fixing this
until IE9. This is the kind of thing that really outlines why they should be releasing updates to their browsers.
Just checked this in IE9 too, and seems that the bug still exists there. Yay for fucking Microsoft.
-
@ASheridan said:
It's not even like the number 4096 makes sense.
Since it's a round number, it makes perfect sense.
There may be reservations about it being so low, or (apparently) unchangeable however.
-
@PJH said:
Since it's a round number, it makes perfect sense.
If you're using 12 bits for your int value... what happened to the other 4 bits?
-
@dhromed said:
Nothing. Someone probably thought (and I use the term in its loosest sense) 0x2000 or above was too large, so they went with 0x1000.@PJH said:
Since it's a round number, it makes perfect sense.
If you're using 12 bits for your int value... what happened to the other 4 bits?
-
@PJH said:
Nothing. Someone probably thought (and I use the term in its loosest sense) 0x2000 or above was too large, so they went with 0x1000.
I just don't understand what thought process would lead to that though? You'd still have to use those bits because they'd be part of the memory allocation no? I can't believe that the developers of IE would come up with a clever way to use only 12 bits of 2 bytes and somehow recycle those other bits for another purpose. It's like the old limit on number of rows in an Excel doc. That I could understand because it was down to the 2-byte limit (65535).
-
@ASheridan said:
@PJH said:
Sorry - didn't make myself clear; I was thinking of an array (or some similar container) with (hard coded) 0x1000 elements, not an integer type using only the lower 13 bits.Nothing. Someone probably thought (and I use the term in its loosest sense) 0x2000 or above was too large, so they went with 0x1000.
I just don't understand what thought process would lead to that though? You'd still have to use those bits because they'd be part of the memory allocation no?
-
Oh, I see. Still, it seems like an odd limit to impose, especially given the proliferation of mobile devices pushing people to combine CSS files in order to speed up their sites load times and reduce server requests.
-
And the problem is obviously with Microsoft and not the webpage that has OVER 4,000 SELECTORS. Obviously.
-
@blakeyrat said:
And the problem is obviously with Microsoft and not the webpage that has OVER 4,000 SELECTORS. Obviously.
OK, I'll bite, but only because I'm bored and actually have nothing better to do.I explained one reason this might happen. I'll explain it once more, and please try to pay attention, I'll use smaller words too so it helps.
Imagine you've just built a large website. This website is huge, lots of pages with complex designs. In this site it's quite easy for a CSS file to grow quite large. Now most developers I know don't split out their CSS on a per page basis, they split it out into logical groups, like one CSS file for standard page elements, another for product pages, another for blog pages, etc. Still with me?
The developer gets another requirement. They have to shave off every millisecond from page load time that they can. One thing they do to achieve this is to combine assets used on the page. So individual images get placed into a sprite graphic, and CSS/JS files are combined and minified. But oh no, combining those 3 CSS files for this group of pages now means that the CSS selector count is above 4000.
Sure, this isn't a frequent scenario, but it could happen. The fact is, there's a limit where there's no reason to have a limit. The limit serves no purpose that I can fathom.
So yes, this is a Microsoft problem. Just like they had that stupid limit on the number of rows you could have in one sheet of an Excel document (something that even OpenOffice, written in Java no less, wasn't prone to). Or are you going to say "And the problem is obviously with Microsoft and not the spreadsheet that has OVER 65535 ROWS. Obviously." to that too?
-
@ASheridan said:
The developer gets another requirement. They have to shave off every millisecond from page load time that they can. One thing they do to achieve this is to combine assets used on the page. So individual images get placed into a sprite graphic, and CSS/JS files are combined and minified. But oh no, combining those 3 CSS files for this group of pages now means that the CSS selector count is above 4000.
Why would shoving all that CSS into a single file save load time? Unless you could guarantee that the visitor will visit every single template on the site, 90% of that shit's going to be wasted. Did you bother to actually benchmark?
I don't believe the scenario of having 4000 selectors in a single file can happen to a sane developer who is working on a sane project. There's no way that can happen without loads of WTF involved.
Look you're shitty at web development. It's ok, you can just say so.
-
@blakeyrat said:
Why would shoving all that CSS into a single file save load time?
If you have to ask, then you're showing you don't know shit about web development yourself. I'll give you a clue, it's all about the time it takes the user agent to make a request to the server and the server to respond, allowing for x amount of concurrent requests. Look you're shitty at web development. It's ok, you can just say so.
-
@ASheridan said:
Look you're shitty at web development. It's ok, you can just say so.
It's true! sobs
No I just don't like seeing mindless Microsoft bashing. Especially IE bashing. It's over now, IE's just as good as all the other browsers, get the fuck over it.
For every idiot going, "oh hey IE doesn't support 4000+ selectors!" there's someone like me going, "oh hey the Firefox developers obviously didn't even fucking READ the DOM specs because the behavior of removeEventHandler is utterly wrong". All browsers are equally shit. I'd even go as far as saying that IE is less shit than the others.
