W3C home > Mailing lists > Public > www-font@w3.org > October to December 1997

OpenType, get it right

From: Clive Bruton <clive@typonaut.demon.co.uk>
Date: Fri, 24 Oct 97 01:19:54 +0100
Message-ID: <1334466365-25890469@battersea.indx.co.uk>
I'm on my way to Redmond, you can all discuss this:

In response to Daniel Will-Harris' article at: http://news.i-us.com/wire/


***************************************


OpenType, get it right, why?

Because Adobe and Microsoft should do their best to get it right, as the only defence type developers have, and Daniel should do some research, or at least take note of what he is told, and leave his bias at home.

The evidence as presented:

>The browserıs new OpenType font embedding feature 

This isn't OpenType, as your own research has shown (below), except you seem to ignore it's meaning and choose to repeat it verbatim.

>Netscape 4ıs ³TrueDoc² dynamic font embedding system 
>from Bitstream does not have this problem.

Is that a piece of personal bias?

The reason TrueDoc doesn't have this problem is that it steals design rather than embedding fonts/outlines/software.

While it's true that the TrueDoc outlines accessed via Navigator are not easy to capture, that isn't true of other TrueDoc applications.

>Embedded OpenType, ³.eot² files, are supposedly only 
>useable inside IE4, and, like Navigatorıs font system, 
>are tied to the domain they were embedded for, so they 
>canıt be displayed on other sites. But IE4ıs current 
>implementation makes these fonts extremely insecure.

You prove the reality of this yourself, but again choose to ignore the facts (below).

>The only TrueType fonts that are protected are those with 
>the ³embedding bit² set to ³No embedding.² 

Read that paragraph a couple of times, do you get it?

It's TrueType fonts we're talking about not OpenType (proven again below), and if the font had been set to "No Embedding" then we wouldn't have seen this story in the first place. This is nothing new, TrueType embedding has been around as long as TrueType.

>This bit allows the type foundry to decide...

I'll use the word "stupid" a lot in the next paragraph or so, just to emphasise how stupid I think this is. Stupid companies may well *not* have set their TT embedding options to No Embedding, that doesn't make them any more or less stupid now than when they first started doing it several years ago. The fact *is* that Non Embeddable fonts don't have this problem, other kinds have for a looooong time.

So who's stupid?

>Currently, the embedding tool only works with TrueType 
>fonts, but Microsoft plans to release a version that 
>supports Type1 fonts by the end of the year.

Which will obviously have to be in an OpenType wrapper, or they won't work, therefore the embedding bits will be accessable to be set.

>Type1 fonts have no ³embedding bit,² ...

Not strictly true, depending on the application PS fonts can contain information to prevent embedding, or conversion to other formats, or anything else. However, just like TrueType Embedding this has to be a cooperative thing, ie applications have to recognise and abide by the rules set.

It just happens that very few applications have been set up with rules to recognise PS embedding "bits".

>Microsoftıs official response to this problem contradicts
>earlier statements about the safety of fonts, as well as 
>shows a serious lack of concern over the matter:

If you read this you'll see that your assessment of it is incorrect:

(DWH quoting Simon Earnshaw)

>³It's widely accepted within the font industry that font 
>embedding mechanisms are never completely foolproof,

I accept that, we just need to try as hard as possible, and continually develop, to make it 99.9% foolproof.

>For this reason Microsoft's approach has been to use a 
>proven, stable technology developed with the help
>and blessing of the font industry. Our embedding 
>technology has been a feature of Microsoft Office 
>applications such as Word and PowerPoint since 1991. 

I'm not sure who blessed it, but *please* take note, "since 1991", OpenType wasn't even thought of in 1991, so how can this be an OpenType problem?

>Microsoft always respects the level of embedding set 
>by the original font creator, 

But other developers may not

>and because we use the actual outlines created by the 
>font vendor, we respect the IPR and skill of the vendor.

ie "we are embedding with permission", we haven't stolen someone else's design.

>This is something that the embedding of mathematical 
>approximations of the design can never guarantee.

ie TrueDoc steals intellectual property rights.

>a commitment demonstrated by our forthcoming OS support 
>for process-private fonts. 

Which will also need support from other OS vendors.

(Back to DWH)

>Microsoft is putting all the responsibility for font 
>security on the part of the font foundries. 

Er, one giant leap for DWH, what embedding bits do is let the foundry *decide* whether the want to support embedding, and at what level, surely that's their decision to make, not Microsoft's (or Adobe's). Should Microsoft ban all embedding at an OS level?

>However, the foundries I spoke with all said they would 
>have set their embedding bit differently had they known 
>about this problem. 

Stupid people do stupid things, the embedding bit's been there all along.

>This would allow vendors to revise the embedding status 
>of their libraries to ³No Embedding,² regardless of what 
>the embedding  bits actually say. 

So now you're suggesting that stupid foundries should tell all their customers that "sorry, we didn't mean that you could *really* embed fonts when we set the embedding bit to "On"" !?

>As a charter member of TypeRight, a non-profit group 
>formed to protect typeface designs,  Iım strongly 
>recommending that Microsoft remove the embedding tool from 
>their site..

Did you ask the other members of TypeRight to endorse your view? I think they might object to your suggestion that TrueDoc is a better solution.

Besides anything else TypeRight is mainly concerned with the issue of copyright on type design in the US, this is not and never was a design copyright issue, it's a licensing and software copyright issue. The former is a problem for stupid foundries, the latter is being fought in the US courts right now.

Furthermore as a member of TypeRight you should be ashamed at your ignorance of the damage done by TrueDoc.

As a founding member of TypeRight I object to your use of our group as an aid to your opinion.

>Copyright Daniel Will-Harris, 1997, All Rights Reserved.

Now a mention of copyright, Daniel protects his words, but forgets about our fonts.
Received on Thursday, 23 October 1997 20:23:32 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:38 UTC