WOFF Only?

Are we at that point in history where we only need to deliver WOFF and WOFF2 web font kits to users? Can we fully abandon .EOT? .TTF? I do believe .SVG can be now ignored.

In part my dream is to also fully remove .TTF or .OTF files so we're only delivering WOFF and WOFF2 . . .

Too soon?
Tagged:

Comments

  • Thomas PhinneyThomas Phinney Posts: 2,730
    You can get away with it pretty much. The most important difference is that WOFF is not supported by Android 2.2–4.3 browsers, while TTF/OTF is supported by those browser versions. They make up all of about 1% of web browsing, though. (About 35–30% of Android installed base, I think.)
  • Previously, apparently: http://typedrawers.com/discussion/1522/webfont-formats
    They make up all of about 1% of web browsing, though. (About 35–30% of Android installed base, I think.)
    How could Android be just ~3% of web browsing?
  • Do you know if .WOFF2 has addressed any of the issues present in the original WOFF?

    With WOFF essentially just being a compression wrapper around .OTF .TFF I believe it's relatively easy to reverse engineer the file to retrieve the font inside.
  • For future reference, you can also consult caniuse.com
  • Jens KutilekJens Kutilek Posts: 338
    With WOFF essentially just being a compression wrapper around .OTF .TFF I believe it's relatively easy to reverse engineer the file to retrieve the font inside.
    It must be easy to retrieve the font, web browsers must be able to do it quickly ;)
  • There are easy ways to convert WOFF2 to whatever: woff2_decompress or ttx. The point of WOFF2 is really awesome compression.
  • John HudsonJohn Hudson Posts: 2,955
    edited June 2016
    The point of WOFF2 is really awesome compression.

    And decompression. The format is tailored to a balance between file size and speed of decompression to get the webfont downloaded and displayed as quickly as possible.

    It's possible to make smaller files than WOFF2 compression, but they take longer to decompress, and the overall download+decompression time ends up being longer, on average, than a slightly larger file that decompresses more quickly. Was very interesting watching the process of determining the appropriate balance, especially since the actual amount of compression varies quite widely across a large body of fonts, determined by a whole range of factors in how the fonts are made.
  • SiDanielsSiDaniels Posts: 277
    Web font DRM would rely on closed proprietary standards, and that's really not on the cards - the web is built on open standards.
  • James PuckettJames Puckett Posts: 1,969
    …the web is built on open standards.

    Or at least it was until people with gobs and gobs of money wanted DRM in HTML video.

  • Richard FinkRichard Fink Posts: 165
    edited June 2016
    …the web is built on open standards.

    Or at least it was until people with gobs and gobs of money wanted DRM in HTML video.

    I'm starting to like you, Puckett. 

    I, personally, don't feel in my gut that it's time to leave out EOT quite yet.  Windows 10 has both IE and Edge and I don't know what the deal is with that yet. And there are still a measurable number of IE8 users out there.

    EOT DOES NOBODY ANY HARM.  It's a Hippocratic thing. 

    As a former network system engineer and network/desktop support person, I trust myself completely on this.  My gut has always worked real good for stuff like this. Sorry to sound like Donald Trump. 

    However, I have been convinced that NOT including the ttf or otf file is best.  Why torture users of old versions of Android with low bandwidth connections?

    If the browser doesn't handle Woff, you might as well let the fallback system fonts do whatever it is they are going to do.  In time, this problem should go away completely. Bandwidth is increasing all the time and the need to buy a new phone or tablet sooner or later, is an inevitability for any user. 

    Last thing - a WOFF2 wrapped version of the font should definitely be inserted in the stack, BEFORE the WOFF. Do it today.  

     I do believe .SVG can be now ignored.

    Yes. Dump it. It was a workaround to accommodate Opera or somesuch and is long obsolete. Chris Lilley of the W3C mentioned exactly this in a video.  If anybody's interested, if a search doesn't bring it up, I'll dig up the link.
  • Had two requests for "legacy" web formats this week. One for SVG (suggested TTFs instead) and one for TTFs. In 95% of the cases, WOFF(2) + EOT will be enough (I imagine clients still expect EOT). But you'll regularly encounter people working for clients that want to cover their base. Of course you can always tell them to use fall-back fonts… if you are okay leaving them somewhat unhappy.
  • Just like habits, old formats die hard. Case in point: officially, Windows XP support ended over two years ago. Every 8th of the month I get a warning about it when I boot... but about every couple of months I also get an automatic update! :-)  So Microsoft is basically being like a mother warning her son to finally move out of the house already... but making him breakfast anyway just to be safe.
  • James PuckettJames Puckett Posts: 1,969
    Had two requests for "legacy" web formats this week. One for SVG (suggested TTFs instead) and one for TTFs.

    Just do it and charge them whatever your minimum custom job rate is. If someone need to spend a thousand dollars for peace of mind just let them do it. In the context of a big marketing campaign this stuff is immaterial. 

  • SiDanielsSiDaniels Posts: 277
    edited June 2016
    Make every glyph in the SVG a middle finger emoji, and see if the client ever complains. ;-)
  • Max PhillipsMax Phillips Posts: 474
    Just literally had a client write, "Thanks for the webfont files. But where's the eots?"
Sign In or Register to comment.