IE 7 does not resize text sized in pixels

Lately I have seen several people mistakenly claim that Internet Explorer 7 allows the user to resize text that has a size specified in pixels, and here is my go at doing what I can to prevent that from spreading and becoming an urban myth.

IE 7 does not, repeat not, resize pixel-sized text. It partly compensates for that by allowing the user to scale/zoom the entire page, including images (which quickly leads to massive horizontal scrolling because of its bad implementation). But if you choose “Page/Text Size” from the menu to change the text size, nothing happens if it has been sized in pixels or an absolute unit (points, millimeters, centimeters, inches, or picas).

Just to make everyone aware of that.

I can’t help wondering why Microsoft is being so stubborn about this.

Posted on March 20, 2007 in Accessibility, Browsers, CSS

Comments

  1. Yeah, I was hoping it might change that too. No such luck.

    Grrr.

  2. Following my incorrect post I’ve been thinking about this a little more. Agreed, there are many resources which state that IE 7 does support text resizing with pixels - Richard Rutter’s on 24 ways being perhaps one of the bigger ones.

    I think what’s causing the confusion though is that the zooming feature in IE 7 uses the same keyboard shortcuts as text resizing in Firefox and Safari.

    Of course what isn’t often mentioned is that Opera doesn’t really offer a text resizing feature at all (other than the ability to set a minimum font size) and through its page zooming feature.

  3. I guess because when something is sized in pixels, the designers/developers want it to look exactly like that? If px and ems behaved the same way, wouldn’t that be illogical? And isn’t it just a bad habit to use px for font-sizes?

    Now, don’t get me wrong, because I think that all websites should have text that resizes - but I think we can already achieve that using ems… Why are we (as developers) being so stubborn about using px?

  4. March 20, 2007 by Fred

    IE’s behaviour is correct IMO. Resizable text should never be specified in absolute pixels. Using em or %

  5. @kimblim: you might think something sized in pixels will “look exactly like that”, but that isn’t what the spec says: a CSS pixel is explicitly defined in such a way that it is not bound to the size of a device pixel, but rather is relative to the output device’s resolution. See CSS 2.1 section 4.3.2 for details.

    I strongly agree with you about using ems for sizing text, though. In my experience, those who insist on using pixels tend to be the kind of designer who think they can achieve the same kinds of control over a user’s screen as they can over the output of a printing press. Before long they start asking questions like “How wide is the scrollbar, so I can get the width right?” and - my personal favourite - “How many pixels high is the content area of the browser window at 1024 x 768?”

    (In the case of that last one I started reeling off a list of the different sizes I could achieve by enabling and disabling various toolbars, moving the address bar next to the links toolbar, switching to fullscreen, and so forth. The hapless designer seemed to find it quite irritating.)

  6. March 20, 2007 by Seyora

    I personally prefer IE’s method of zooming over that of Firefox. Their “scale zoom” keeps the proportions of all the page elements so it doesn’t break the layout. (But if I recall correctly, Opera can also zoom in like that though, so IE still doesn’t win. Heh.)

  7. March 20, 2007 by Ryan

    @Fred

    Please explain exactly why you think resizeable text should not be specified in pixels?

  8. I always begin with percent to stop IE’s text resize bug which comes to life if you specify ems on body. I then use ems or percent throughout my style sheets as needed. Thus, due to IE7’s lack of resize support, I don’t have to modify my practice.

    For the record, pixel sizing is not absolute. Points and picas may be but pixels are not. That is a common misunderstanding. Pixels are a relative unit even though IE screws it up.

  9. Well.. I am confused, is it a bug in MSIE or right behaviour?

  10. I do have some sympathy with Microsoft’s stance, assuming it is a stance and not an oversight.

    While pixels may be a relative unit they are not relative to browser settings. My understanding of pixels, while admittedly incomplete, leads me to expect that type set in pixels will be sized the same way as any graphics, but that setting type relative to the browser default will allow it to be resized by the browser itself.

    In many ways then IE is actually more flexible. It is very easy to use relative font sizing (and I do) so I don’t really see the problem with having an option in the arsenal that allows us to choose non-relative font sizing. As most of us use pixels to size our graphics it would seem to make sense that pixels are the non-relative unit.

  11. I personally hate the zoom :(

    I know when I first showed a client how you should increase the text twice to see if it breaks and I tried it in IE and got a huge surprise when I saw the image enlarge

  12. Pixel values are safe if you like to create a well-defined “standard feeling” for a website - if you have a designer’s heart, you will love pixel values. Any good browser will resize these values if the user wants to do that. Of course your pages should not break then. But using “em” or “%” values means to do s.th. just because there is IE.

    IE is not the web, nor will it ever be.

    IE does not resize text. So what? People who are visually impaired should not use IE, and they will not. I hope all the others will change their browser, too. (Some weeks ago I was impressed about the “new” IE (7) - digging deeper I learned that it has the same bad core than IE 6.)

    Meanwhile I do not care anymore about text resizing in IE. Maybe we all should use pixel values - in the end we would help people to change to better browsers. Ignite the change, don’t do everything for IE!

  13. One new design that seems to have caused some confusion is Happy Cog’s. Dan Mall’s font size technique for Happy Cog was to use pixels for all browsers except IE; IE’s style sheet has font sizes in “ems”. Dan’s comments are here.

  14. Because Microsoft only goes after billion dollars deals, not million dollar deals.

    Internet Explorer does not have to be perfect to be a success.

  15. @Nick Fitzsimons: I must admit that I did not read the spec for a css pixel, but if it is relative to the resolution, surely resizing text shouldn’t change the size of a pixel? Changing your screen resolution should? Before anyone starts bashing me, I am not pro or anti IE, merely curious to why they have adopted the stance? And if they might be right? Maybe we as developers/designers just got spoiled when other browsers made it possible to resize text set in pixels?

    @Everyone: I don’t know if you have seen this screencast, but I can only recommend that you take the time to do so. Apart from Håkons constant IE bashing (which has its time and place), it is quite interesting - especially listening to Chris Wilson explain the issues that the IE team deals with.

  16. I don’t care about ie - it’s getting down and user’s screens are getting bigger - not so much people use function of text resizing.

    Not everybody even knows about this feature, mostly those who use alternative browsers.

    ie7 can rescale - that’s it. If you don’t like hos it works - use better browser. You use beeter browser? So, you don’t have problems with text resizing…

  17. As Fred already mentioned, the IE behaviuor is correct. It may be not very useful, but it is correct. If a web designer decides to fix some text to a certain pixel size, than this text should have exactly that size in pixels. It is what the designer wanted. Imagine navigation buttons with an image of a button as background image. If the text on these buttons would be resizable this would break the layout. I had a longer discussion about that in the Mozillazine forums. I learneds there that this “Feature” of the Firefox was put in only because of those many badly designed web pages, where socalled designers fixed the font size in pixels, and this was unreadable on other screens. So Firefox adds a convenient way to at least make those pages readable. This feature is a necessary evil, but not correct behaviour. Unfortunately it’s needed urgently, but at least you cant’t blame microsoft for an erroneous implementation here. It would be better if IE had that feature, too, but correct is correct. The correct place for sizing text is in the settings. There you can set up the default 8standard) text size. Firefox has this, and IE has it too. It’s just that too many of those socalled web designers do ignore what the user has set up. It is even common practice to at first set the standard text size to something about 60%, just because the preset default text size is 16px, which many designers think is way too large. But just think of what this means: A person who really needs a larger text and would be content with a font size of 16px has to set a much larger text size to get at least these 16px. It is like the web designer decides: “You won’t get 100%, i’ll give you only 60%”. Well, at least these types of font sizing is correctable by the user. Fixing the font size to pixels is only correctable by ignoring the standards, which much too often is necessary and much too often then breakes layout.

  18. March 21, 2007 by Roger Johansson (Author comment)

    Siegfried:

    If a web designer decides to fix some text to a certain pixel size, than this text should have exactly that size in pixels. It is what the designer wanted.

    But it may not be what the user wants, and that’s what is important here. On the Web, users have control. IE’s behaviour is completely backwards.

    Imagine navigation buttons with an image of a button as background image. If the text on these buttons would be resizable this would break the layout.

    <irony>Oooh, disaster! The user can increase text size so he/she can read the text. And it makes my pretty design somewhat less pretty! The horror!</irony>

    A responsible Web designer will take care to make sure everything can scale when possible, and when it is not possible make sure that text is still readable even if the graphics surrounding it don’t match perfectly anymore. When I need to increase text size, I am doing it to get at the information. I do not care whether everything still looks perfect. If it does, great, but it is unimportant.

  19. March 21, 2007 by GreLI

    Roger Johansson:

    A responsible Web designer will take care to make sure everything can scale when possible…

    But if designer do so there is no need in pixels. Just use ems or percents. So browser must not resize pixel sized text and IE behavior is correct. If some text is unreadable address it to a designer not to browser.

    Pixel sized text is not necessary small text. It may be large logotype that must have room in layout and if you’ll make it even more larger the page became too big.

    Now only Firefox resizes text sized in pixels and it isn’t standart. Just another original browser implemantation (remember Netscape? Mozilla is it descendant). Some browsers can set minimum font size though. And this makes more sense in case of small text size.

  20. March 21, 2007 by Niels

    If users have the control, it’s time for the W3C to abandon px sizing.

    And until they do so 1. we just continue developing sites in ems and nothing’s the matter and 2. browsers should stick to the standards.

    You know what happens when browsers go against the standards. Let’s be fair to them (and ourselves) and at least believe the stuff we preach.

  21. Roger, I completely agree with you when you say “It is not what the user wants”.. I’m just sure it’s the opposite thing the IE team said when they released all those IE versions with all their non-standard stuff, that some developers could use to make the pages “cool”.

    The problem here, as I see it anyways, is that there is no clear standard: Should text set in pixels be resizeable or not? If there indeed is a standard like Nick wrote, that defines a pixels size relative to the device that outputs it, IE has got it right. Unfortunately that standard is not very user-friendly.

  22. March 21, 2007 by Roger Johansson (Author comment)

    GreLI:

    But if designer do so there is no need in pixels. Just use ems or percents.

    Of course. But since most designers do not care about the end user, users have to be able to compensate for that by being able to resize text even if it is specified in pixels or an absolute unit.

    Now only Firefox resizes text sized in pixels and it isn’t standart.

    Every browser except for IE does it. Well, Opera zooms, but at least it does it in a reasonable way.

    kimblim:

    The problem here, as I see it anyways, is that there is no clear standard: Should text set in pixels be resizeable or not?

    Right. So what is left is what is to be sensible about it. And that is to let all text be easily resizable by the user, no matter what unit is used. End of story as far as I am concerned.

    Everyone: To anyone claiming that IE is right about this, I’d appreciate if you could point me to some document that clearly states that user agents are not allowed to resize text whose size is specified in pixels or an absolute unit. Show me such a document and I’ll admit that IE is right. It won’t change my mind about what is common sense though.

  23. March 21, 2007 by Pete

    What’s most useful for the user is the main thing, which is being able to resize text set in pixels. The horizontal scaling in page zooms look pretty unusable to me, so text-sizing(on any unit) should be the norm.

  24. IE7 also “thinks” that fieldset is an inline element. Why we even talk about that dumb browser?

  25. I agree that IE should allow the user to override the designer’s decision and resize the text, even if it was specified in pixels, but…

    I’d appreciate if you could point me to some document that clearly states that user agents are not allowed to resize text whose size is specified in pixels or an absolute unit.

    Is there a document that states the opposite?

  26. I agree that IE should allow the user to override the designer’s decision and resize the text, even if it was specified in pixels, but…

    I’d appreciate if you could point me to some document that clearly states that user agents are not allowed to resize text whose size is specified in pixels or an absolute unit.

    Is there a document that states the opposite?

  27. Roger, I agree that we should be sensible about it, but to me, that means using ems, not pixels. As long as there is no standard, we can’t tell if IE or Firefox is implementing the right solution - and from previous bad experiences, I do not want to develop for a specific browser.

  28. I’d just like to add that when I wrote in my 24 Ways article that IE7 could resize text spec’d in pixels, I was implying the page zoom feature. I stand by that assertion (in as much as readers can make the text bigger) although I apologise to readers if it lead to some confusion.

    I’d also like to add that, in an ideal world, I believe sizing text in pixels is the most appropriate method for text that will be read on a screen. Ems are an alternative, but I would prefer to reserve ems for specify lengths in relation to text size (margins etc). However while we are stuck with IE6 (and arguably IE7) sizing text in pixels in inadvisable due to the resizing problem. Which is a bug, not a feature.

  29. Sorry my comment was posted twice, but I got a server error the first time!

  30. Everyone: To anyone claiming that IE is right about this, I’d appreciate if you could point me to some document that clearly states that user agents are not allowed to resize text whose size is specified in pixels or an absolute unit. Show me such a document and I’ll admit that IE is right. It won’t change my mind about what is common sense though.

    CSS 2.1 Section 18.5 Magnification

    User agents may support such magnification in different ways

    When magnifying a page, UAs should preserve the relationships between positioned elements. For example, a comic strip may be composed of images with overlaid text elements. When magnifying this page, a user agent should keep the text within the comic strip balloon.

    Sounds like IE7 behaves appropriately to me…

    The ‘Text Size’ preference in IE is used solely for changing the default font (font-size: medium) and has nothing to do with resizing text.

  31. The picture element (px) are relative to the viewing device whereas ‘em’ is relevant to the font. Thus pixels are relative and are allowed to be resized or rescaled. In contrsat absolute length units are only useful when the physical properties of the output medium are known and hence in many cases they are just approximates.

  32. March 21, 2007 by GreLI

    Pixels must not be resizable by definition. If you resize them (except zoom — when you get all resized) they are not real pixels. When 1 pixel became 1.5 pixel while other 1 px is still 1 px — that is nonsense. I think zooming and minimum font size are enough to fight small text size. So resizing elements in pixels is excessive and an absolutely wrong way.

    Look what happening: we say that designers must make they output resizable. But because they are bad and sometimes don’t do that then we’ll totally change font size (given in pixels) in browser. But this is bad way too. If someone makes bad things it is not reason to do bad things too.

    And if a browser have some bad implementations this does not mean that it is tottaly bad and developers are dumbs. They may have sensible reason to do things in they own way.

  33. you are not doing bad things, you are helping users to cope with the stupid designers. unfortunately, IE seems to be stupid too.

  34. That IE behavior is 100% correct. RTFM.

  35. March 21, 2007 by Roger Johansson (Author comment)

    CG: Thanks for a very articulate and civil response. Care to point to a more specific place where everybody can RTFM?

  36. IE is indeed correct here. A pixel size is a physical size, depending on the physical size of a pixel on the media. This size is by nature not resizeable. Resizing page text, or just the necessity of resizing page text, is bad. It should not be necessary. The default text size, which the user selects, should be the text size the user gets. Nothing else. It is bad habit to enforce the user to resize the text of every page. The Firefox feature is just there because this bad habit is enforced by bad web designers. The only correct way to size the text is in the browser settings and by the user, nothing else. Never! And there are scenarios where a font size in pixels is necessary. These should be rare exceptions and only for small parts of the web page, but still there are those scenarios. And only for these scenarios there is the possibility to fix a font size in pixels. And just because some ignorant doubts my web design skills: My own pages do not have a single font size fixed in pixels. My own pages do not need them. Just for info. The resizing feature of the Firefox is a necessary evil, needed because of too many bad designed web pages. But that does not mean, because we have that feature we have to design bad web pages! I personally would prefere that the Firefox team eliminates this feature: It leads to bad web design. Font sizes in em or percent is not just there because the IE cannot resize pixel sized fonts. That’s complete nonsense. Font sizing relative to the default (standard) font is the normal way to go. For any browser. Font sizing relative to the physical constraints of the media (aka pixel size) is not recommendable and should be used with caution and rare.

  37. I thought this had been an issue that was fixed by the IE team for IE7, I was surprised when I found this out the other day.

    The zoom function may be a more useful way to enlarge the text but I think it has a lot of issues regarding usability. Particularly people going between different browsers will find this behaviour difference confusing. Then again I find IE7’s interface confusing all round!

  38. @Siegfried:

    A pixel size is a physical size, depending on the physical size of a pixel on the media. This size is by nature not resizeable.

    This is only true if you take the term “pixel” to mean “device pixel”. CSS, since version 1, has explicitly stated that a CSS pixel is not the same thing as a device pixel. CSS 2.1 gives a list of relative units, and they are em, ex and px. A single CSS pixel is to be mapped to however many device pixels may be necessary to satisfy the viewing requirements given in the spec. CSS does not support the concept of a device pixel.

    Dave Hyatt (ex-Mozilla, now working on WebKit at Apple) wrote an article last year explaining the nature of a pixel and other units in CSS, in which he states that “Those of you who think (a pixel is a screen pixel) have a fundamental misunderstanding regarding how the Web works”. It’s a good read, and the preceding articles about high DPI devices go into some detail about the techniques needed to address the requirements of the CSS 2.1 spec.

  39. I think they may just like being in the “pain in the ass” gang. Giving users to resize pixel text might be too conformist for them?…only the IE team hold the answers…

  40. March 21, 2007 by GreLI

    Ideally designers should use pixels when it is important to place it correctly among other pixel sized elements: such as images or may be layout (feel free to think your own variants). And resizing text without other elements (i. e. without complete zooming) like specification says (see comments above) is incorrect.

    Particularly one can say that images must resizable too in ideal (i. g. when desktop monitors will have something like 300 dpi), but even than ideal solution for backward compatibility will be zooming not just text resizing.

    So resizing text with size given in pixels is a bad solution which was made in order to get better usability of some bad designers output.

  41. March 21, 2007 by Ingo Chao

    Text Size does not resize the text in IE.

    The user was clicking on that menu entry and nothing happened. But he wanted to see this text larger. He thought it was intended for sizing text, it was printed on the label.

    Computer programmers like to gain control over a machine. Users would like to have some control over an application. What’s the position of the designer?

    The user doesn’t care about what a designer thinks a px stands for, or if the designer thinks it’s a bug or not. For the user, it does not work. Who has the final say?

  42. I’m a bit concerned by the repeated referencing of a spec as if that should be the last word on everything. That’s not to say I think it should be outright ignored (whether it states px should be resizable or not), but if designers want to use px and users expect text to be resizable, shouldn’t that ultimately be what determines the UA behaviour? And shouldn’t that in turn inform “the spec”, at least the common understanding of how it should be implemented? If IE7 doesn’t conform to a convention that every other browser adheres to, and that users rightly expect, isn’t that just as bad as not conforming to the spec?

  43. There are several different aspects of this discussion interweaved:

    • The specs on text sizing.
    • The functionality of the browsers.
    • User’s knowledge and ability to change their browser.

    I suppose you could see pixels as relative to other pixel sized elements, therefore increasing pixel sized text should be part of a function that increases everything else that’s in pixels. This would also increase graphics, but depending on the page layout, might not increase the layout, or might just increase parts of it.

    How browsers should increase sites when you could be using a combination of percentages, pixels and font based units is incredibly complex. It is no wonder that Opera opts for one-method and does a good job on that one, and IE7 has had a hard time with it’s implmentation.

    The main thing is that the definition of text size should be separate from the content (no one’s arguing with that, but I think it’s worth making explicit).

    When that is done, people can resize the text. In IE you have to dig through the preferences and ‘ignore font size settings’, for other browsers this function is built into the main interface.

    The difference could be seen as cosmetic from a web-dev’s point of view, but the approach has ramifications:

    • If that IE setting is adjusted, then all sites text are ‘abnormally’ large, the difference is not subtle, and will often break layouts. This has made it a more extreme option that isn’t used much.
    • Not allowing pixel-fonts to resize with the view functionality (until IE7’s zoom), means that the function behaves erratically from a user point of view working on some sites but not others.

    IMHO text resizing functionality has never been given a chance to gain widespread adoption due to IE’s implementation.

    For those who said that people should not use IE if they need to resize text: a lot of people are stuck on locked-down corporate networks where they cannot install anything else. I’ve met a few people at usability testing in large corporate and government organisations who either didn’t know you could change the text size for some sites, or weren’t allowed to. Removing the system setting is probably illegal under the DDA in the UK (still untested), but it still happens.

    I don’t think IE7s zoom has improved the situation much. It’s zoom is an improvement for fixed-width (pixel oriented!) sites, but it needs to mature to get to Opera’s level of usability across different sites.

    There was quite a good discussion on browser’s zoom on accessify.

  44. March 21, 2007 by Zephyr

    I agree that developers and web designers really should start using em instead of px. But since this is unlikely to happen on a big scale anytime soon, I’d rather give people the power to easily enlarge text through their browser. It almost seems MS gets some perverse pleasure from evading standards.

  45. It’s very interesting that we as a group disagree as much as we do. That is probably because there are no clear standards defined on the issue, and if that is true, we don’t know (yet) whether IE or Firefox handles text resizing the right way.

    Maybe it’s good (short term) that the browsers handles these things differently, because it will give the W3C some test cases to actually determine which is the better solution.

  46. March 22, 2007 by Isaac Lin

    For those using resizable pixels as defined by the W3C standard as a reason to support resizing text specified in pixels, I agree with comment 40: going just by the standard, scaling should happen when the entire pixel unit is scaled up, including the graphics, such as is done with IE7’s page zoom feature (leaving aside for the moment whether or not this feature should re-layout the page based on the larger pixel sizes). It is inconsistent to use the standard as a justification for scaling the text pixel unit independently from the pixel unit used for all other elements.

    That aside, browser implementors are free to implement additional extensions to the user interface that are useful. A best of both worlds would be to support pixel resizing and text resizing.

  47. I think I found CG’s FM to read, the user agent accessibility guidelines.

    In a nutshell:

    Ensure that the user can select preferred styles (e.g., colors, size of rendered text, and synthesized speech characteristics) from choices offered by the user agent. Allow the user to override author-specified styles and user agent default styles.

    I don’t see an exception for px, or any other unit of measure. Magnification or zoom seems to be acceptable.

    cheers,

    gary

  48. March 22, 2007 by John W

    Anyone who things that pixels are absolute units are flat out wrong.

    PIXELS ARE NOT ABSOLUTE UNITS. In CSS2, they are in fact relative units that relate to the users’ environment. This is particularly important when designing for High-Resolution/HD screens in the future as pixels afford the designer a great deal of consistency and accuracy with layout positioning and can be zoomed without any issues.

    A great description of why pixels are considered relative units can be found here:

    http://webkit.org/blog/?p=57

    Yes, the post also talks about why ems are useful:

    “The other units worth discussing are em and ex. These units are interesting to use primarily because of the text zoom feature of Web browsers. Because some browsers (like Safari and Firefox) just zoom text without zooming other content, expressing everything in these units can give you a way of having content other than text also scale with the font size.”

  49. March 22, 2007 by Niels

    It is inconsistent to use the standard as a justification for scaling the text pixel unit independently from the pixel unit used for all other elements.

    I really agree with this. If you want to make px a sizeable unit (mind that sizeable is different from relative), be consistent and size everything defined in px.

    Why would you want to size fonts but not other elements on a site ? It might be easier, but is deliberately breaking a design something that we should applaud ?

  50. March 22, 2007 by Roger Johansson (Author comment)

    Why would you want to size fonts but not other elements on a site ? It might be easier, but is deliberately breaking a design something that we should applaud ?

    Because it is much easier for the end user to handle in the browser window. When everything zooms it tends to lead to very wide pages very quickly, which in turn leads to horizontal scrolling, making orientation on the page more difficult. I often need to increase text size, and I much prefer increasing only text instead of zooming.

    It is not deliberately breaking a design since design for the Web should be able to handle increased text size as well as different amounts of content. Bad designs break quickly, good designs break later.

  51. March 22, 2007 by Isaac Lin

    Resizing the pixels doesn’t have to lead to wide pages with a reasonable width browser window if the browser recalculates the page layout when the pixels are resized. Of course, the placement of any fixed-width pixel elements (typically images) will dictate a minimum pixel width, and so eventually with a large enough pixel size, there will be horizontal scrolling.

    Text placed within images (like the example given of balloon help) is probably the main issue with resizing the text only. Sliding doors techniques could be used to cope with plain backgrounds, but for something like text in dynamically generated maps, only resizing the text would cause a problem.

    As I noted, I am not objecting to text-only resizing, but am only saying that those who are using the W3C definition of a resizable pixel as a reason should support page zoom.

  52. March 22, 2007 by Chris

    Hi all. My name is Chris and i am on a project “em based design”. This needs a lot of research. So i am glad Roger wrote this article. I want to say, Roger, i have become you’re biggest fan. Your’re totally awesome!

    Roger, i totally AGREE with you on this point. I want your opinion about this, read my comment very carefully, especially the last part is where its all about…

    I know yourself DIDN’T used my solution at this site, because you didn’t need to. But you, as an usability expert did it wrong. Why? Your menu at the top of the page with ‘Home’, ‘About’, ‘Archive’, ‘Lab’, etc… is at the almost TOP of the html code.

    For usability aspects (especially for screen readers & google optimalising) you have to put the Menu at the bottom of the page. People with screen readers wants to read text, not menu’s.

    But when you make use of Posistioning, its difficult to do it wright with Em’s for cross-browser ha?

    All those people claiming to be “professional developers” are wrong. Actually, they are robots. They are programmed to do the things they’ve learned. The programmers told the “PD’s” that pixel based fonts is WRONG, so they will never think different again.

    PD’s(“professional developers”): Go do some research, don’t learn the books, but practise MAKING websites again.

    Step 1: Make a complex website based on EM’s. Measure almost everything in EM’s. Yeah, even all divs, paddings, margins, positionings etc etc. Only the area’s around images can be pixeled (see comment 40) Step 2: Set the font-size in the BODY-TAG in % or EM. Step 3: Make it work Cross-browser.

    What is my point? Well, this CANT work, because of the EM. Why not? Firefox for Windows has a different EM measurement then Firefox for MAC and Safari is again totally different from firefox for MAC. So the cross-browser part will NEVER work.

    Solution? It’s simpel, i TOUGHT i came up with it myself, it took me some time to get it work. After i got it work i saw this article + comments, and Very suprisely….

    Comment #13. 2007-03-21, 3.28 by Sean Fraser is excactly the solution!!!

    So the problem is:

    -Modern browsers acts rare on EM’s -Internet Explorer don’t resize on pixels.

    Solution: -Set the Body TAG font-size at modern browser at PIXELS!!! -Set the Body TAG font-size at Internet explorer at EM’s!!!!

    NOW you can specifie all other TAGS in EM (like Div p span h1 margins, paddings, line-heights, ETC ETC..

    (except for the area around images, read comment 20)

    ET VOILA it works! Both the TEXT and the BOX around the text Resizes like it should be. Only thing to do is set the max-width on some EM’s so you don’t get horizontal scrol-bars.

    So PD’s, dont think pixel-sized text font-size in the BODY TAG is wrong for modern browsers, because of the pixel-sized text in the body, all modern browsers get the same web-site-layout.

    IE is wrong in this. BODY TAG pixel-sized text font-size SHOULD be resizable, if the rest of the tags are in EM’s mainly.

    But for this moment, we need to specify a different font-size for Internet explorer in the body tag in %.

    conclusion: Modern browsers: body{11px/1.4em Verdana} Internet Explorer: body{68.75%/1.4em Verdana}

    Thank you for reading, sorry for my bad english.

    Roger, what is your opinion about this?

    Greetz, Chris.

  53. Isn’t there a difference between magnification and text resizing? I work with people who have visual disabilities and they all say that there is a difference. Magnification is like a magnifying glass, it magnifies a portion of the page, where you want to read. My co-workers and friends all like elastic designs that limit the width. This allows them to make the text large enough to read but keeps the text in context to the page. Unfortunately I haven’t yet set my site to limit width on text resize.

  54. Personally i couldn’t care less about IE and its inability to resize font-size set in pixels.

    When i receive a design from our designers it’s so much easier to simply measure everything in pixels (in Photoshop) and put the exact same thing in the CSS. Font-size, margins, paddings whatever.

    Ok, i’ve never really learned em’s and that, but why not use pixels?
    Like someone else said, if you’re serious about resizing the font-size while surfing, don’t use a crappy browser like IE.

    I reckon font-resizing should be available always, regardless of what’s set in the stylesheet.
    IE is only stubborn, lazy or stupid.

  55. March 23, 2007 by Niels

    Because it is much easier for the end user to handle in the browser window. When everything zooms it tends to lead to very wide pages very quickly, which in turn leads to horizontal scrolling, making orientation on the page more difficult. I often need to increase text size, and I much prefer increasing only text instead of zooming.

    Well, when the design doesn’t zoom, it leads to insane vertical scrolling. You seem to like that and find it useful, personnaly I’d rather scroll sideways a bit and be sure to read more than one word per line.

    For this particular issue, I think it’s important to what visually impaired people (as a group, not as individuals) have to say about this. Because what’s easier for one person, is a whole lot more difficult for another.

    Loosely related, but are there any fora / blogs / places where one can gather opinions of the people we’re all so concerned about ?

  56. March 23, 2007 by Roger Johansson (Author comment)

    Because what’s easier for one person, is a whole lot more difficult for another.

    Yes, and that is exactly why having both options would be the best solution for the end user. It would be great to see Gecko and WebKit implement a zoom feature (as long as they make it work more like in Opera than in IE). As long as they don’t do that, however, resizing text regardless of which unit is used is much better than offering nothing, like IE 6.

  57. March 23, 2007 by Niels

    As long as they don’t do that, however, resizing text regardless of which unit is used is much better than offering nothing, like IE 6.

    Well, we got into this mess because browsers tried to correct developer’s mistakes, remember. I’d rather that browsers stick to the standards. At least that would put some pressure behind the people developing these standards. Now browsers will fix their inconsistencies and errors. Which lead to very little progress the last 6 years, and to a heap more problems.

    I’ve been doing some comparisons between the IE7 zoom and Opera zoom myself. All I can say is IE seems to provide something closer to an actual zoom, while Opera does something in between which starts breaking my layout again and messes up simple graphics where IE doesn’t. Both of them seem to have a problem with balance (fe, one border will jump to 2px, while the others remain at 1px).

    I prefer IE7’s more visual zoom instead of Opera’s clunky “who knows what it’s doing” zoom.

  58. Well, there is a problem with using ems in IE. This occurs with high definition laptops, most of which are set to 120 DPI. IE will display “medium” text at 20px at this setting. I find this breaks an awful lot of sites.

  59. March 24, 2007 by Roger Johansson (Author comment)

    Chris: Thanks :-)

    For usability aspects (especially for screen readers & google optimalising) you have to put the Menu at the bottom of the page. People with screen readers wants to read text, not menu’s.

    Opinions on that vary: Source Order, Skip links and Structural labels. And there are other people to cater for than screen reader users. Keyboard users, mobile phone users, text-only users, etc. So I don’t think you can claim that something like the main navigation on this site should not be before the content in the source.

    But when you make use of Posistioning, its difficult to do it wright with Em’s for cross-browser ha?

    It is more complex than with pixels, but it can be done.

    Firefox for Windows has a different EM measurement then Firefox for MAC and Safari is again totally different from firefox for MAC. So the cross-browser part will NEVER work.

    I don’t believe this is correct. There are minor differences in some cases because of rounding errors (what one browser rounds to 11 screen pixels may be rounded to 12 pixels in another browser), but as far as I am aware browsers do not have different em measurements.

    Roger, what is your opinion about this?

    I am not so sure I like the idea of sending different font sizes to different browsers. It reminds me too much of what we used to do back in the nineties for Netscape. But the method you are suggesting is still worth taking a closer look at.

    Niels:

    Well, we got into this mess because browsers tried to correct developer’s mistakes, remember. I’d rather that browsers stick to the standards.

    I see what you mean, but I don’t see how allowing the user to resize text can possibly be a problem. Besides, allowing that is the standard, as pointed out by Gary Turner in comment #47. I don’t see any mention of user agents not being allowed to resize text depending on which unit is used in the UAAG. The UAAG does state, however, that zoom or magnification is a sufficient technique, so I guess that leaves a loophole for IE.

    My opinion is still that not allowing text resizing regardless of unit is very user-unfriendly.

  60. March 25, 2007 by Chris

    Roger:

    I have read that article. I must respect your opinion for this site. But aren’t you focused too much on the screen readers / other like devices? You should consider this:

    • Screen Readers vs SEO

    Russ Weakley:

    Source order is often irrelevant as most screen reader users and refreshable Braille device users have a variety of ways that they can move around pages quickly (skipping to heading levels, links, forms, form elements, as well as tabbing quickly through content and doing direct searches for content on the page). The concept of “top” and “bottom” of pages often becomes irrelevant.

    Fact:

    Having content ordered at top (before main navigation) increases the SEO of a page.

    If your audience is mainly the least experienced screen reader user you are right about the main navigation on top.

    I’m not very experienced on mobile phone users, (could you give me a link were I can find info on that + testing) but I think it depends on the mobile browser / the content of the menu (are there only 5 items, or are there like 15 menu items & 100 sub menu items)

    In most situations the SEO thing is of more value then the screen reader user etc.

    Roger:

    It is more complex than with pixels, but it can be done.

    lol, this part was not mentioned for you, but for other people who are to lazy using positioning in combination with em’s, because they think its to difficult.

    Roger:

    but as far as I am aware browsers do not have different em measurements

    Well, I can say to you, there is a slightly difference in em measurement between Firefox for MAC and Firefox for Windows. Or could you say to me what the difference in potition occur?

    I have an example here (it is only an example, not a real website, only for testing purposes and not html strict valid) witch uses positioning:

    You should look at the difference between Firefox MAC and Safari (witch display the same as Firefox Windows and Internet Explorer to safe you starting parallels).

    Look at the difference between the boxes at the left and right sides in the given browsers. In the code you should look at this (i changed the example here slightly, because the real example uses bad chosen font-sizes in the body):

    body{font:68.75%/1.4em Verdana,Helvetica,sans-serif;}
    #zoeken{position:absolute; top:175px; margin-top:2em; padding-bottom:5px;}
    #log{position:absolute; top:175px; margin-top:2em; padding-bottom:5px;}
    #subnav{position:absolute; top:25.8em;}
    #prev{position:absolute; top:29.6em;}
    

    zoeken and log are the 2 upper boxes, and subnav and prev the lower once.

    If i changed the code, by this:

    body{font:11px/1.4em Verdana,Helvetica,sans-serif;}
    

    and made an apart style sheet for IE, so in IE it is still:

    body{font:68.75%/1.4em Verdana,Helvetica,sans-serif;}
    

    Now all browsers display the same. I have no example for that on this, but I have one for a real, still in progress, project of my…

    Roger:

    I am not so sure I like the idea of sending different font sizes to different browsers. It reminds me too much of what we used to do back in the nineties for Netscape. But the method you are suggesting is still worth taking a closer look at.

    Well, lets take a closer look! I am now at a real project in progress making use of that. I would like your opinion about that, this site makes use of navigation in the bottom of the source, is html valid, and works cross-browser (well not tested on all the browsers you’re testing, but on Firefox Mac, firefox Windows, Safari, Opera, IE 7, 6, 5.5). I also made use of some javascript for the left menu.

    The font size for Internet explorer is body{font-size:68.75%} And for modern browsers: body{font-size:11px}

    The site url for the project in progress is:

    pokertips in progress

    It is only the homepage; no other pages yet, will be continued.

    Ps1. This is a real project, the goal of this project is: change the existing Site:

    real pokertips site

    It must look like the same, but the code must be html / css strict valid and much better for SEO and users with bigger fonts selected

    Ps2. the site is also in a lot of other languages, like

    swedish pokertips site

    Greetz chris.

    Ps3. We could discuss this in private, I don’t want to spam your nice little site with this I am sorry for the hugh posts :$ you’ve got my email, damn I spended 3 hours on this post.

  61. March 25, 2007 by Chris

    but as far as I am aware browsers do not have different em measurements

    forget about it.. there is no difference at all. it was just me who acts very dumb.

    The fault i made was putting the font-size in wrong ems. By making it 0.9em it is 0.9 x 16 = 14.4 pixels. Because of this Firefox MAC acted different from Firefox Windows…. Since that point i believed there was a difference and tried to adjust it… i never tested the firefox in whole rounded pixels in percentages or em.

    so you are so right on this lol:

    There are minor differences in some cases because of rounding errors (what one browser rounds to 11 screen pixels may be rounded to 12 pixels in another browser)

    at the pokertips site, i changed the 11px at modern browsers to 68.75%, and there was no difference at all.

    sorry for me being wrong // it took me some bad spended time, but i’ve learned from it thanx

    Leaves us at the screenreaders vs SEO discussion…

    OKE and some on topic discussion, since im still awake…

    Roger post 22:

    users have to be able to compensate for that by being able to resize text even if it is specified in pixels or an absolute unit. Every browser except for IE resizes text sized in pixels.

    This is true, if you press Command +

    But what about a user who sets a font-size in the Preferences of his browser? (because its get frustrated typing command + every site again…)

    Then, text specified in pixels would NOT resize in ALL browsers.

  62. March 25, 2007 by Roger Johansson (Author comment)

    chris:

    Wow, long comments! Thanks for spending so much time on this :-). I’ll try to answer your questions.

    But aren’t you focused too much on the screen readers / other like devices?

    I like to think that I am browser/technology agnostic in my approach to accessibility. What I mean by that is that I don’t like making special workarounds for a particular browsing device (though in some cases it is inevitable). So no, my reasoning for having the top level navigation (the complete navigation on this particular site) first in the source is not that it will make the site easier to use for screen reader users. I do find that, as a person who likes using his keyboard to navigate and spends a reasonable amount of time browsing the Web from a mobile phone, some navigation is nice to have before the main content. Not if it consists of a huge list of 100 links, but if there are fewer than, say, 10. In combination with skip links I find that to be a good compromise.

    When tabbing through a page, I find it extremely disorienting if a menu bar like the one on this site is visually displayed first, but once you start tabbing it comes after the content. That could be compensated for by using tabindex attributes, but I prefer keeping everything on the page in a logical order.

    With regards to SEO, I have not seen any negative results from doing this, but I am not an SEO expert, so maybe somebody has experimented and has facts. From what I userstand, search engines are quite good at detecting blocks of navigational links and are able to exclude them when going through a site’s content. I don’t know how much truth there is to that, but it does sound plausible.

    Regarding what is most important, it would always be the user for me. Not screen readers or SEO.

    I am now at a real project in progress making use of that. I would like your opinion about that

    Increasing text size works fine in Safari. I do get a horizontal scroll quite quickly though, so perhaps you could do something about that. Other than that, resizing works great.

    But what about a user who sets a font-size in the Preferences of his browser? (because its get frustrated typing command + every site again…) Then, text specified in pixels would NOT resize in ALL browsers.

    Correct, but in that case I would suggest also checking (or unchecking) the option that allows or disallows the site to specify font size at all.

  63. March 25, 2007 by Chris

    Increasing text size works fine in Safari. I do get a horizontal scroll quite quickly though, so perhaps you could do something about that. Other than that, resizing works great.

    I’d looked at it and tried some changes but i’m not sure if it’s better. I’d rather have a horizontal scrollbar then having 3 words a line. You have your point with the horizontal scroll bar, i have my point of the vertical increasing space, i think it will be in between both. Is there some research on this subject?

    I would suggest also checking (or unchecking) the option that allows or disallows the site to specify font size at all.

    I didn’t know that option is available. I looked for it. there is not such an option, however, there is the option of the minimum font-size. But that has completely nothing to do with our discussion:

    • If i set an minimum font-size of 24px, all text will be 24 pixels.
    • if i set a font-size of 24px, 1em = 24px, 0.75em = 18pixels. (but 12pixels will stay 12 pixels etc.)

    I think such users will go for the second options: setting the font-size at 24px and not the minimum font-size.

    Again on this: is there some research done?

    users have to be able to compensate for that by being able to resize text even if it is specified in pixels or an absolute unit.

    They can’t compensate when it is in pixels, they could only adjust all the text to 1 bigger size. They have no opportunity having bigger headers then paragraphs when the text is set in pixels.

    But with ems they have the opportunity of scaling the text in the correct proportion, pixels don’t give that.

  64. A good website has a good fontsize, if some people can’t read them they must go to the opticien or buy a screen reader… But I will keep this in mind when I design!

  65. Now that sucks! grr @ Microsoft. It’s funny how they were the first company to include CSS support in their browser and it’s taking toooo much time for them to understand that people hate ‘em and their browser because of poor CSS support.

  66. At least IE7 can zoom. As for me, I’m done baby-sitting Microsoft’s sorry attempts to produce a browser. We can only coddle the weakest link for so long, until it’s time to just say - Goodbye, laggards.

  67. I like that Firefox can resize only text!

    It keeps the layout of the page as intended by the designer, and in certain cases lets me read easier the text, if I want to increase font size (in case it’s too small, for example).

    I do not like any ZOOM features. Images get scaled as well and they look pretty ugly then.

    IE7 doesn’t allow text resize for websites which specified the text in pixels, and that’a bad, IMHO.

    Lots of designers specify text in pixels still. This is bad practice, considering the fact that the market is still 80% IE vs 20% other browsers.

    If I can’t read Verdana 10px then what do I do? In Firefox, CTRL & +. In IE6? Nothing. I can close the website. In IE7? Zoom. But zoom creates horizontal toolbar, distorts images, and looks pretty ugly.

    So, for me, Firefox is right, and Microsoft IE7 is wrong.

    Let’s hope version 8 of IE will fix this “pixel-defined font sizes” finally:-D

  68. I’m on-board with Nathan Smith here. The accessibility problem has been solved. IE7 can increasae the size of text set in pixels. You can make the distinction between text zoom and page zoom all you want, but the bottom line is that you can increase the size of text set in pixels in IE7. Period. You just can. No more accessibility problem.

    The fact that IE7’s implementation doesn’t make resizing the text as convienent as Firefox or Safari’s doesn’t matter. Accessibility and convienence are not the same thing. DOn’t confuse them,.

  69. April 5, 2007 by Roger Johansson (Author comment)

    Jeff: I get your point, but having to scroll both horizontally and vertically can certainly be an accessibility issue, and it is definitely a usability problem. Jakob Nielsen has more on that in Scrolling and Scrollbars.

  70. I see how it can be a usability problem, but I’m still a bit confused on how it can be called an accessibility issue. To make something accessible is to make it able to be accessed. It’s not to make it able to be accessed without the need or scrollbars. If it was, 90% of desktop apps wouldn’t be accessible, as they almost all require horizontal scrolling at some point. I have five programs (besides my browser) open now: Finder, iTunes, Photoshop, and TextMate. Each of them have a horizontal scroll bar showing. Does that make them inaccessible? I don’t think so. In fact, the sole purpose of those scrollbars is to make me able to access content that doesn’t fit on the screen. The entire reason scrollbars exist is to make things accessible.

    In the real world, a building is considered to be accessible to wheelchair-using individuals if it provides ramps. That’s because the ramps make the building able to be accessed by these people. No one complains that the building aren’t accessible because they require ramps in order to use. Why do we say something is inaccessible if it requires scrollbars to use?

    When I was in music school there was a blind guy in my jazz band. We blew up the sheet music really freaking big with a copier. In order to read this music, he had to put his face very close to the music. This meant he had to move his head — not just his eyes — in order to scan along the page. What this inconvenient for him? Probably. But was it accessible? Absolutely.

    It’s downright impossible to give everyone the exact same experience. That’s not the goal of accessibility. The goal of accessibility to make everything able to be accessed. Text sized in pixels is accessible in IE7, because it’s able to be accessed — even if folks have lower vision need to zoom in on it first. Just as people in wheelchairs might have to take a different route in order to be able to access a building or a blind individual might have to move his head more than the rest of us in order to read sheet music, so might someone with lower vision need to zoom in before the can read type on a screen. This is an inconvenience, and that’s unfortunate. But that content is accessible. It’s able to be accessed.

  71. April 5, 2007 by Roger Johansson (Author comment)

    I see how it can be a usability problem, but I’m still a bit confused on how it can be called an accessibility issue. To make something accessible is to make it able to be accessed.

    Accessibility is not just about making content possible to access. It goes beyond that - it is also about making things usable for people with disabilities. See the difference?

    Having to scroll both horizontally and vertically may be just an inconvenience for a non-disabled person (though I would call it a severe inconvenience). For a person with a disability that makes it harder to use the Web in the first place it is definitely more than just an inconvenience.

    If IE 7 had implemented a better zoom feature that prevented (or at least delayed the point at which it happens) massive horizontal scrolling, we wouldn’t be having this discussion :-). Opera is better, but it isn’t perfect either.

  72. Accessibility is not just about making content possible to access. It goes beyond that - it is also about making things usable for people with disabilities. See the difference?

    I definitely see the difference, but I’m not sure I agree. Either way, it doesn’t matter. We should, of course, strive to make things as usable as possible for as many people as possible. I would never disagree with that.

    But, like all things, there are tradeoffs. The math of doing complex designs with relative units is quite complicated, and results in a lot more time and effort being put into them. Sometimes, this trade off simply doesn’t make good business sense.

    I recently built a huge, complex site (BoomerGirl.com) using all relative units — even the images and layout are in relative units. It worked great, and I felt like it was necessary because the target audience for that site was an older group that was more likely to have vision issues. However, the tradeoff was that the CSS for that site took me far longer than anything else I’ve ever done. It was a lot of work, and a lot of complicated math. In that case, it was well worth it — but that’s not true for every site.

    As for the page zoom feature itself — I actually like it. I would personally rather all text resizing worked that way at all times. I figure if someone has trouble seeing small text, they’ve got trouble seeing small images, too. And I like that it always keeps the layout in tact, no matter how good or poor a job the web designer might have done. I think this page zoom functionality is the way browsers are heading (look at the iPhone, for example) — so we should probably get used to it. :)

  73. April 7, 2007 by Colin Wyers

    What I want to know is — if a CSS pixel is not related in any way, shape or form — why in the hell is it called a pixel? I mean, things wouldn’t be so confusing if the CSS spec didn’t decide to create their own definitions for words already in common usage.

  74. April 7, 2007 by steve faulkner

    “IE 7 does not resize text sized in pixels”

    I am I missing something here?

    In IE 7 as in IE 6 and prior versions, the user can check the “ignore font sizes…” in the “internet options > accessibility” dialog. Then the IE text resize function works regardless of the type of unit used to set font size. They sure don’t make it easy, but it’s not impossible.

  75. April 8, 2007 by Chris

    The Opera zoom function is better then the IE, because of the horizontal scroll-bar issue in IE.

    Jeff Croft:

    I figure if someone has trouble seeing small text, they’ve got trouble seeing small images, too. And I like that it always keeps the layout in tact, no matter how good or poor a job the web designer might have done.

    You can’t say because of bad web designers (you mean developer?), zooming is a good function. It will NEVER be a good function, but it will ALWAYS be there. So it’s up to the user rather he uses the zoom or text-resize (i call it the css resize) function. The zoom function can be used at bad coded sites, but it is an advantage to use the css resize function because the developer can have control over it (yes even resize images).

  76. April 8, 2007 by Roger Johansson (Author comment)

    Steve:

    In IE 7 as in IE 6 and prior versions, the user can check the “ignore font sizes…” in the “internet options > accessibility” dialog.

    Yup, correct. It has (always?) been possible, but very few people know about that option. What I am referring to is the default settings.

  77. What I am referring to is the default settings.

    I’m curious how much responsibility you think falls on users to know how to use their software (and hardware, for that matter). Because the reality is that if we could count on people to know how to use their tools, this would be a non-issue. There are a million choices for someone who needs to be able to resize text. Among them: switch to Firefox, change your settings in IE, and use the built-in accessibility zoom on your computer.

    I’ll again reference the real world here. People with unique needs are, in the real world, expected to obtain and use additional tools to accomodate those needs (see glasses, magnifying glasses, wheelchairs, crutches, and prosthetics for examples). Designers, in the real world, are not expected to make their products work exactly the same for these users, but they are expected to make sure they play nice with these additional tools (for example, architects much ensure that wheelchair ramps are of a reasonable incline steepness).

    I have neck and back problems, myself. Because of that, I have to buy a decent chair and take the time to configure it for the optimal support. If I don’t, I’ll end up in pain when I’m reading a book at my desk. If I choose to sit in a crappy chair, or I choose not to configure my chair the way that is best for me, or I simply don’t learn how my chair works — whose fault is that?

    I’d say it’s my fault. You can maybe make an argument that it’s the chair manufacturer’s fault, because they didn’t make it easy enough for me to use my chair properly.

    But I don’t see any way in hell you can blame it on the book I’m reading. Do you?

    This is why I beleive that more responsibility needs to be placed on the users with special needs, as well as the manufacturers of the tools (computers, browsers, input devices, etc.). We as web designers need to do what we can — but we can’t do it all. These other two parties need to share the responsibility, as well.

  78. April 10, 2007 by Roger Johansson (Author comment)

    This is why I beleive that more responsibility needs to be placed on the users with special needs, as well as the manufacturers of the tools (computers, browsers, input devices, etc.). We as web designers need to do what we can — but we can’t do it all. These other two parties need to share the responsibility, as well.

    I agree fully. I am not saying that you absolutely cannot use pixels to size text. However, when you do that, (well, even if you don’t, but especially if you use pixels) I think it would be a good thing to do what you can to inform your visitors that they do have a choice, and that their software actually can resize the text provided that they change the default setting. Or they can use another browser.

    IE is unfortunately a special case because of its market share. If any other browser was as broken as IE, we could just ignore it, but the number of IE users is forcing us to cover up for what Microsoft is unable or unwilling to do. It sucks, and if there was any justice in this world, Microsoft should pay part of our salaries.

  79. It sucks, and if there was any justice in this world, Microsoft should pay part of our salaries.

    Amen, brother. :)

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.