pixel font sizes

Discussion in 'HTML' started by fefewf, Sep 29, 2007.

  1. fefewf

    fefewf Guest

    why is using pixel font sizes wrong?
    fefewf, Sep 29, 2007
    #1
    1. Advertising

  2. fefewf wrote:
    > why is using pixel font sizes wrong?
    >
    >

    Because for folks who use MSIE that cannot change the size. If a visitor
    cannot read the text...bye-bye visitor.

    If you use em to size your fonts, it's like your library having
    large-print versions of every book. This simple practice allows you to
    offer a large-print version of your site.

    --
    Take care,

    Jonathan
    -------------------
    LITTLE WORKS STUDIO
    http://www.LittleWorksStudio.com
    Jonathan N. Little, Sep 29, 2007
    #2
    1. Advertising

  3. Scripsit Jonathan N. Little:

    > fefewf wrote:
    >> why is using pixel font sizes wrong?


    I wonder whether we being trolled.

    > Because for folks who use MSIE that cannot change the size.


    Actually, there is one mistake and one misleading point in that statement,
    although it's generally a good answer to people asking stupid questions.
    Oops... there are no stupid questions, just... But to the point:

    1) People using Internet Explorer (officially called Windows Internet
    Explorer now, though some people prefer using the older full name or its
    abbreviation) _can_ change font sizes on web pages. Most of them just don't
    know how. (They need to use a setting that overrides _all_ font sizes set on
    web pages, or use a user style sheet with !important.) Even fewer also care
    to, so the basic conclusion is indeed:

    > If a visitor cannot read the text...bye-bye visitor.


    2) The misleading part is that the formulation suggests that this is some
    kind of browser peculiarity, rather than the way browsers are required to
    behave. If you set font size in pixels, then pixels shall it be. This is one
    of the few things that IE implements by the book and many other browsers
    don't. Letting font size increase affect such sizes is comparable to having
    a control that lets the user specify the size of a millimeter or the
    duration of a second. (In fact, the size of millimeter, inch etc. _do_
    change if the monitor resolution is changed, but this is a different
    oddity.)

    > If you use em to size your fonts, it's like your library having
    > large-print versions of every book. This simple practice allows you to
    > offer a large-print version of your site.


    Except that this also happens if you don't set font size at all. However,
    setting font-size: 100% is recommended (though it is dummy in principle) as
    a weapon against some browser bugs. Using % is somewhat less risky than
    using em for font size, again due to browser bugs.

    --
    Jukka K. Korpela ("Yucca")
    http://www.cs.tut.fi/~jkorpela/
    Jukka K. Korpela, Sep 29, 2007
    #3
  4. Jukka K. Korpela wrote:
    > Scripsit Jonathan N. Little:
    >
    >> fefewf wrote:
    >>> why is using pixel font sizes wrong?

    >
    > I wonder whether we being trolled.
    >
    >> Because for folks who use MSIE that cannot change the size.

    >
    > Actually, there is one mistake and one misleading point in that
    > statement, although it's generally a good answer to people asking stupid
    > questions. Oops... there are no stupid questions, just... But to the point:
    >
    > 1) People using Internet Explorer (officially called Windows Internet
    > Explorer now, though some people prefer using the older full name or its
    > abbreviation) _


    WIE, or should I say WEE! Whatever. I like MS's naming that lends to the
    already entrench misconception that MS(W)IE is required to browse the
    Internet on Windows system.

    > can_ change font sizes on web pages. Most of them just
    > don't know how. (They need to use a setting that overrides _all_ font
    > sizes set on web pages, or use a user style sheet with !important.) Even
    > fewer also care to, so the basic conclusion is indeed:


    Most IE(okay?) users are ignorant enough of the menu "View > Text Size>
    ...." let alone fool with user stylesheets. Sorry just been my observation.

    >
    >> If a visitor cannot read the text...bye-bye visitor.

    >
    > 2) The misleading part is that the formulation suggests that this is
    > some kind of browser peculiarity, rather than the way browsers are
    > required to behave. If you set font size in pixels, then pixels shall it
    > be. This is one of the few things that IE implements by the book and
    > many other browsers don't.


    Agreed. Although it is small consolation on what they miss... but hey,
    if all browsers followed a common spec where would be the challenge in
    web design?

    > Letting font size increase affect such sizes
    > is comparable to having a control that lets the user specify the size of
    > a millimeter or the duration of a second. (In fact, the size of
    > millimeter, inch etc. _do_ change if the monitor resolution is changed,
    > but this is a different oddity.)


    Agreed, discrete units px, in, cm, ... should not be user adjustable.
    But that said, their implementation in webpages' visual style should be
    carefully considered. The ramifications of bad design would be more
    severe if browsers like Firefox didn't break the rules, been to some
    sites that would not be accessible at all!

    >
    >> If you use em to size your fonts, it's like your library having
    >> large-print versions of every book. This simple practice allows you to
    >> offer a large-print version of your site.

    >
    > Except that this also happens if you don't set font size at all.
    > However, setting font-size: 100% is recommended (though it is dummy in
    > principle) as a weapon against some browser bugs. Using % is somewhat
    > less risky than using em for font size, again due to browser bugs.
    >


    But, it the bug in question concerning that MS browser, and occurs if
    you use em units to set the base font size? If you set the base font
    size with % you can reference off of that in em's without problems?

    body { font-size: 100%; }
    ..legalese { font-size: .75em; }

    --
    Take care,

    Jonathan
    -------------------
    LITTLE WORKS STUDIO
    http://www.LittleWorksStudio.com
    Jonathan N. Little, Sep 29, 2007
    #4
  5. "Jukka K. Korpela" <> writes:

    > In fact, the size
    > of millimeter, inch etc. _do_ change if the monitor resolution is
    > changed, but this is a different oddity.


    Just to clarify that point - obviously, the real-world size of these units
    does not change. What changes is the number of pixels per inch your monitor
    is displaying. Sadly, many browsers(1) simply assume a 96ppi display, and
    fail to adjust for reality.

    So, if you specify a 1in margin (for example) in CSS, what you'll usually
    get is 96px - which will only result in a 1" margin if your monitor is set
    to exactly 96ppi. If your monitor is 72ppi instead, you'll get a 1.25"
    margin, and at 120ppi, .75".

    sherm--

    (1) Mozilla is one exception - its control panel has a pane where you can
    hold a ruler up to the screen to calibrate the PPI setting it uses when
    translating physical units into pixels.

    --
    Web Hosting by West Virginians, for West Virginians: http://wv-www.net
    Cocoa programming in Perl: http://camelbones.sourceforge.net
    Sherm Pendley, Sep 29, 2007
    #5
  6. fefewf

    dorayme Guest

    In article <ScvLi.230722$>,
    "Jukka K. Korpela" <> wrote:

    > In fact, the size of millimeter, inch etc. _do_
    > change if the monitor resolution is changed, but this is a different
    > oddity.


    I'd like to see that! (But I know what you mean).

    --
    dorayme
    dorayme, Sep 29, 2007
    #6
  7. fefewf

    dorayme Guest

    In article <>,
    Sherm Pendley <> wrote:

    > (1) Mozilla is one exception - its control panel has a pane where you can
    > hold a ruler up to the screen to calibrate the PPI setting it uses when
    > translating physical units into pixels.


    I recall iCab being another.

    --
    dorayme
    dorayme, Sep 29, 2007
    #7
  8. fefewf

    cwdjrxyz Guest

    On Sep 29, 8:34 am, "fefewf" <> wrote:
    > why is using pixel font sizes wrong?


    As several others have pointed out, various browsers respond to pixel
    font sizes in somewhat different ways and for different screen
    dimensions, thus other methods for controlling text size often work
    better for a wide range of browsers. One extreme way to avoid many of
    these problems is to make the whole page as an image, including the
    text. This actually has been done, but usually for reasons other than
    making the page more viewable. Some who put up pages or send out email
    that is likely to get blocked if certain words are included sometimes
    use just an image for the questionable material. Most filters for
    blocking content do not scan words in images. Of course you can set
    your browser not to show images but then you may block images you
    might wish to see, especially if alt for the image is set to something
    misleading making you think you might wish to turn on images. Making
    too much of the page as a large image may greatly increase the file
    size over using text and a few small images, if required. This would
    sometimes slow your page down too much if you are aiming for viewers
    who have only a slow dialup connection. I see no need for making most
    of the page an image for my personal use. However, some spammers and
    scam artists apparently think it is of benefit to them to do so. They
    must think that they will gain more viewers by avoiding being blocked
    by scans of text than they will lose from viewers who have images
    turned off.
    cwdjrxyz, Sep 30, 2007
    #8
  9. fefewf

    Bergamot Guest

    Sherm Pendley wrote:
    >
    > (1) Mozilla is one exception - its control panel has a pane where you can
    > hold a ruler up to the screen to calibrate the PPI setting it uses when
    > translating physical units into pixels.


    FYI, that feature never worked on Windows.

    --
    Berg
    Bergamot, Sep 30, 2007
    #9
  10. fefewf

    - Bob - Guest

    On Sat, 29 Sep 2007 14:25:36 -0400, "Jonathan N. Little"
    <> wrote:

    >Most IE(okay?) users are ignorant enough of the menu "View > Text Size>
    >..." let alone fool with user stylesheets. Sorry just been my observation.


    Exactly the point. Developers debate this font issue endlessly - most
    users don't even realize IE has menus. I've never met a non-developer
    who knew what a sytlesheet was. Too many techies worry about picayune
    minutia that matters only to other techies.

    The same rule applies to page width arguments. A business web site is
    a marketing tool for business, you want it to have the appearance you
    design it for - not to have a technical design that makes techies
    drool. Suggesting that it needs to resize endlessly to any browser
    width is like suggesting that an architect's building design should be
    able to be optically stretched to fit on any lot... just let all the
    windows, doors, walls, plumbing, heating, etc all stretch to a smaller
    or larger lot size.
    - Bob -, Oct 3, 2007
    #10
  11. fefewf

    dorayme Guest

    In article <>,
    - Bob - <> wrote:

    > Suggesting that it needs to resize endlessly to any browser
    > width is like suggesting that an architect's building design should be
    > able to be optically stretched to fit on any lot... just let all the
    > windows, doors, walls, plumbing, heating, etc all stretch to a smaller
    > or larger lot size.


    What about the suggestion that it needs to be able to stretch a
    bit, contract a bit, fit on a goodly number of different screens
    and platforms and browsers, be reasonably comfortable for a range
    of people of different capacities? What's this like? A house that
    has good insulation, the doors open and close without jamming...?

    --
    dorayme
    dorayme, Oct 3, 2007
    #11
  12. Scripsit - Bob -:

    > On Sat, 29 Sep 2007 14:25:36 -0400, "Jonathan N. Little"
    > <> wrote:
    >
    >> Most IE(okay?) users are ignorant enough of the menu "View > Text
    >> Size> ..." let alone fool with user stylesheets. Sorry just been my
    >> observation.

    >
    > Exactly the point. Developers debate this font issue endlessly - most
    > users don't even realize IE has menus. I've never met a non-developer
    > who knew what a sytlesheet was. Too many techies worry about picayune
    > minutia that matters only to other techies.


    People who really, really need font size control (such as visually impaired
    people, who might prefer visual browsing even though e.g. 6opx is the
    smallest font size they can read) do know about browser menus and other ways
    to user control over font size. They might even have a user style sheet,
    perhaps written by a helpful techie friend.

    To such people, fixed font size in pixels is not an issue because it does
    not exist.

    Fixed font size in pixels (or points or millimeters) is a problem just to
    the few billions of people who do not absolutely need font size control but
    cannot read the typical tiny font sizes comfortably (or at all). And they
    often simply leave the problem by going elsewhere. If you wanted to sell
    them something, you just told them to ask your competitors.

    P.S. By "tiny", I mean effectively anything below 10pt. For readability on
    paper, the font size of copy text should be between 10 and 12 points; that's
    a reasonable compromise between conflicting needs, and the exact value
    depends on the font, but generally 12 or 11.5 is more suitable than 10. On
    screen, the font size should generally be _larger_ than on paper, since the
    presentation medium is coarser and the reading conditions vary more. So even
    12pt should be regarded as acceptable only if it is easily controllable by
    users, as far as it depends on the author.

    Magically, this means that the best thing you can do to copy text font size
    is doing nothing. Well, except for font-size: 100%, which is logically a
    dummy setting but helps to avoid some bus.

    --
    Jukka K. Korpela ("Yucca")
    http://www.cs.tut.fi/~jkorpela/
    Jukka K. Korpela, Oct 4, 2007
    #12
  13. fefewf

    Vaxius Guest

    On Sep 29, 1:25 pm, "Jonathan N. Little" <>
    wrote:
    > Jukka K. Korpela wrote:
    > > Scripsit Jonathan N. Little:

    >
    > >> fefewf wrote:
    > >>> why is using pixel font sizes wrong?

    >
    > > I wonder whether we being trolled.

    >
    > >> Because for folks who use MSIE that cannot change the size.

    >
    > > Actually, there is one mistake and one misleading point in that
    > > statement, although it's generally a good answer to people asking stupid
    > > questions. Oops... there are no stupid questions, just... But to the point:

    >
    > > 1) People using Internet Explorer (officially called Windows Internet
    > > Explorer now, though some people prefer using the older full name or its
    > > abbreviation) _

    >
    > WIE, or should I say WEE! Whatever. I like MS's naming that lends to the
    > already entrench misconception that MS(W)IE is required to browse the
    > Internet on Windows system.
    >
    > > can_ change font sizes on web pages. Most of them just
    > > don't know how. (They need to use a setting that overrides _all_ font
    > > sizes set on web pages, or use a user style sheet with !important.) Even
    > > fewer also care to, so the basic conclusion is indeed:

    >
    > Most IE(okay?) users are ignorant enough of the menu "View > Text Size>
    > ..." let alone fool with user stylesheets. Sorry just been my observation.
    >
    >
    >
    > >> If a visitor cannot read the text...bye-bye visitor.

    >
    > > 2) The misleading part is that the formulation suggests that this is
    > > some kind of browser peculiarity, rather than the way browsers are
    > > required to behave. If you set font size in pixels, then pixels shall it
    > > be. This is one of the few things that IE implements by the book and
    > > many other browsers don't.

    >
    > Agreed. Although it is small consolation on what they miss... but hey,
    > if all browsers followed a common spec where would be the challenge in
    > web design?
    >
    > > Letting font size increase affect such sizes
    > > is comparable to having a control that lets the user specify the size of
    > > a millimeter or the duration of a second. (In fact, the size of
    > > millimeter, inch etc. _do_ change if the monitor resolution is changed,
    > > but this is a different oddity.)

    >
    > Agreed, discrete units px, in, cm, ... should not be user adjustable.
    > But that said, their implementation in webpages' visual style should be
    > carefully considered. The ramifications of bad design would be more
    > severe if browsers like Firefox didn't break the rules, been to some
    > sites that would not be accessible at all!
    >
    >
    >
    > >> If you use em to size your fonts, it's like your library having
    > >> large-print versions of every book. This simple practice allows you to
    > >> offer a large-print version of your site.

    >
    > > Except that this also happens if you don't set font size at all.
    > > However, setting font-size: 100% is recommended (though it is dummy in
    > > principle) as a weapon against some browser bugs. Using % is somewhat
    > > less risky than using em for font size, again due to browser bugs.

    >
    > But, it the bug in question concerning that MS browser, and occurs if
    > you use em units to set the base font size? If you set the base font
    > size with % you can reference off of that in em's without problems?
    >
    > body { font-size: 100%; }
    > .legalese { font-size: .75em; }


    There's no base font-size here: 100% of what?

    >
    > --
    > Take care,
    >
    > Jonathan
    > -------------------
    > LITTLE WORKS STUDIOhttp://www.LittleWorksStudio.com
    Vaxius, Oct 5, 2007
    #13
  14. Vaxius wrote:
    > On Sep 29, 1:25 pm, "Jonathan N. Little" <>
    > wrote:


    >> body { font-size: 100%; }
    >> .legalese { font-size: .75em; }

    >
    > There's no base font-size here: 100% of what?


    the BODY is the base of the displayed document, text outside of the BODY
    element is not presentational.

    body { font-size: 100%; }

    means 100% of the "user defined | browser default" font size, all other
    displayed elements are children of the BODY and are referenced off of
    its settings...

    --
    Take care,

    Jonathan
    -------------------
    LITTLE WORKS STUDIO
    http://www.LittleWorksStudio.com
    Jonathan N. Little, Oct 5, 2007
    #14
  15. fefewf

    Vaxius Guest

    On Thu, 04 Oct 2007 20:46:09 -0400, Jonathan N. Little wrote:

    > Vaxius wrote:
    >> On Sep 29, 1:25 pm, "Jonathan N. Little" <> wrote:

    >
    >>> body { font-size: 100%; }
    >>> .legalese { font-size: .75em; }

    >>
    >> There's no base font-size here: 100% of what?

    >
    > the BODY is the base of the displayed document, text outside of the BODY
    > element is not presentational.
    >
    > body { font-size: 100%; }
    >
    > means 100% of the "user defined | browser default" font size, all other
    > displayed elements are children of the BODY and are referenced off of
    > its settings...


    Nice, I didn't know you could do that.
    Vaxius, Oct 5, 2007
    #15
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. coleenholley
    Replies:
    5
    Views:
    1,127
    coleenholley
    Jan 21, 2004
  2. BGW

    Static Font Sizes

    BGW, Jul 7, 2003, in forum: HTML
    Replies:
    6
    Views:
    3,019
    Matthias Gutfeldt
    Jul 8, 2003
  3. mcp6453

    Font Sizes

    mcp6453, Aug 10, 2003, in forum: HTML
    Replies:
    4
    Views:
    496
    Nico Schuyt
    Aug 11, 2003
  4. André Wagner

    Fastest pixel by pixel operation

    André Wagner, Mar 17, 2008, in forum: Java
    Replies:
    12
    Views:
    1,027
    Larry A Barowski
    Mar 19, 2008
  5. mttc
    Replies:
    2
    Views:
    2,365
    Roedy Green
    Jul 3, 2009
Loading...

Share This Page