W3C home > Mailing lists > Public > www-font@w3.org > July to September 2009

RE: the discussion is over, resistance time

From: Sylvain Galineau <sylvaing@microsoft.com>
Date: Fri, 3 Jul 2009 00:23:05 +0000
To: Thomas Lord <lord@emf.net>
CC: Gustavo Ferreira <gustavo.ferreira@hipertipo.net>, "www-font@w3.org" <www-font@w3.org>
Message-ID: <045A765940533D4CA4933A4A7E32597E020BEF8E@TK5EX14MBXC111.redmond.corp.microsoft.com>


>> 1. I do not believe the cost of implementing any of the proposals I've
>seen justifies this argument.
>
>We ought to consider the cost of implementation,
>the cost of maintenance, the costs in terms of
>reliability (new code assures new bugs), the costs
>in terms of taking more memory for more code especially
>on smaller / lower-power devices, the costs to users
>left puzzling over a new format and what works with
>what and how to convert things, the costs to archivists
>who must preserve additional tools for digital materials,
>and so forth.
>
>The costs are highly diffuse, certainly, but they
>add up.
>
>If standards orgs regularly adopted your attitude here,
>they would quickly become useless.
I have worked in standards group and I still do. If you want a check
on my 'attitude' on their usefulness, ask Hakon.

>
>
>> 2. Ascender's original proposal, for instance, would be trivial to
>implement.
>
>Implementation is only a small fraction of the
>costs being imposed on the society.


If you mean that these costs should be compared to the cost of the status quo, I agree.
I'm glad we agree implementation is very cheap, at least.

>
>
>> 3. Even if the solution was non-trivial, the combined expense on
>browser vendors is thoroughly dwarfed by the aggregate cost
>> on all web authors and web sites of dealing with the current situation
>where they have to serve each font in two encodings,
>> never mind the lost opportunity of using certain fonts for all
>browsers. The costs to web authors and the consequences for all users
>> should come before the inconvenience of browser vendors, imo.
>
>The costs of serving two formats, the lost opportunities
>and so forth: those look from here to be costs imposed
>solely by Microsoft's intransigence on this issue.


If the cost is low, so are the number of lost opportunities. As for Microsoft, I doubt we'll
be sued for a monopoly on intransigence in this matter anytime soon. Please. :)

>
>Capitulating to that intransigence would damage W3C
>and, I don't know about you, but I think that would be
>the highest cost in the list.

The W3C is proposing the creation of this Working Group, afaik. We're interested. If you're not,
I don't think anyone forces you.

>
>
>> 4. As for all the other programs that may want to use web fonts, we
>are not exactly short of excellent free libraries allowing
>> anyone to create, read and write every single format browsers support
>today e.g. JPEG, GIF etc. I don't see why this couldn't happen
>> for fonts. Your argument could as well have been made for PNG and any
>number of other formats. I don't see why a trivial font encoding
>> is any different in this respect, or how it suddenly pushes the world
>beyond some line of bloatedness. Never mind who defines bloat or how.
>
>"Bloat" occurs over time by the accretion of many
>needless non-features which accomplish poorly something
>that could better be accomplished in simpler ways
>or in more general ways.   Perhaps my comment above
>about the hidden costs of Microsoft's position are
>helpful to you or perhaps not.


The needless non-feature on offer is the same your own proposal aims for.

That you believe your proposal to be superior to all others is well understood but that is not enough to make it so.
That's not standardization works. "If standards orgs adopted your attitude here, they would quickly become useless"

>
>It is especially important to resist the incremental
>accretion of bloat in standards processes because their
>reach is so far and so lasting.   Of particular
>significance is that standards organizations play
>an "iterated game," so that a sloppy compromise in
>one standard sets a precedent that can be exploited
>to insist upon sloppy compromises in future standards
>from the same org.  It's easy to get into a "race to
>the bottom".


Your proposal does not add any less bloat than Ascender's. The latter only deals with fonts.
You seem bent on addressing just about any resource. Somehow, dealing with any type of licensed media
results in less bloat and will be easier to agree on. I don't buy it. I know I'm not the only one.
I suggest you moderate your ambition and deal with the matter at hand. I don't think there is a quorum here
who wants to tackle all licensing on the web, or believes it's that simple. Font licensing is hard enough as it is.

>> 5. Lastly, since your own proposal would involve the exact same kind
>of change by all software that wants to use web fonts, one would think
>it would be as bad an idea
>> as any proposal that requires new code.
>
>My proposal does not involve "the exact same kind of change".
>The only thing in common is that my proposal would create
>a file format, for files containing fonts, where that format
>is not initially recognized by most programs.


Which requires these programs to write new code to handle it. By our own argument, this is unacceptable.

>
>Some key differences are that my proposal would add a
>very general and extensible bit of new utility that is
>not more simply achieved by other means.  Comparing and
>contrasting the content and quality of the rationale
>statements that would go with the respective proposals
>is a worthwhile thought experiment.
>

I'm not saying looking at it is not worthwhile. It doesn't mean it's radically different and can be asserted
to be an OK change while all others constitute bloat.

>> As for font vendors, it is up to them to state whether their goals are
>satisfied by this or that proposal.
>
>
>My perception here - my concern - is that I think I sense
>Microsoft digging in its heels here purely reflexively
>and in search of some kind of validation of its self-conception
>of power and authority.

Well, you're entitled to your opinion. Your concern, however, is just that: an opinion.

>It looks from here like y'all are internally asking one
>another "how do we feel about this TTF/OTF stuff" and people
>are reflexively answering, in essence, "they can't make us
>but we can prevent them, so 'No.'"  It may be dressed up
>in rhetoric about protecting font makers but calmer and more
>logical examinations of that "protection" have shown it
>to be an illusion so I'm left with the sense of a power play
>being made for the sake of asserting who, in Microsoft's
>opinion, is boss.

Font makers do not want raw TTF linking. Some of their representatives
have said so on this mailing list. Others have done so in public in the past.
Lastly, WebKit has supported raw font linking since late 2007. How many commercial font vendors have
licensed fonts allowing for raw linking since then ? It's fine to hypothesize it's all rhetorical, but
it'd be more effective and relevant if you could back it up. So I'm not sure who's dressing up what here.

>I'm sure that isn't Microsoft's official position or even
>the conscious intent of you, or Chris Wilson, or most others.
>That's just what the net effect of your collective reasoning
>comes across as.

That it comes across that way is no proof that is so.

>And I'm sure you *could*, if you wanted, respond by saying
>"Hey, buddy, check yrself!  You are just trying to have the
>other browser implementers dictate to MSFT and show them
>that they're boss!"   But no, that wouldn't hold up either
>in the sense that existing technical landscape and the
>technical details of the competing proposals are quite
>asymmetric - and "our" side has the far more rational set
>of proposals by all measures other than the political.


That may be so but speaking of what people are coming across as, you're doing a poor job of
representing your argument rationally, imo. I can follow Hakon, John Daggett, Tab Atkins and others.
You, on the other hand, lose me. But I'm stubborn and I believe you mean well or I wouldn't bother.

>I don't know, at this point, how to extend an "open hand
>rather than a closed fist" to Microsoft on this matter.
>I *thought*, even perhaps a week ago, that serious progress
>was being made and we were starting to get close to wrapping
>this thing up with a productive outcome.   Now, at this late
>date, it looks like you guys are just plugging your ears,
>shouting "LA LA LA", and taking things back to the status
>quo ante.   Only the status quo ante isn't necessarily
>available since quite a few people have spent a lot of time
>and effort in this area on the assumption that Microsoft
>might actually behave reasonably.

You can't just assert what's reasonable and I don't think you've done a good job explaining how reasonable is measured and why.
Starting with your whole 'bloat' argument. You have failed convincing me at least - and maybe Tab - that this
Reasoning does not apply to your own proposal.

'Reasonable' is a relative criteria that may also still produce different outcomes. Mozilla, for instance, does same-origin check on web fonts by default.
The WebKit team decided that wasn't necessary. You might find both more 'reasonable' than Microosft but the difference matters to the authors who will deploy
fonts for use in both browsers.

Lastly, all of my posts here explicitly state that the status quo is bad for the web. I have no idea how you got the idea that the status quo ante,
where only Microsoft implemented web font support, is anyone's goal including Microsoft's.

But if by status quo ante you meant undoing what's already been released, I agree it's likely not longer available even if it's fair to ask whether raw font linking is seeing significant adoption in practice, whether an alternative could achieve much wider adoption, and whether the short-term migration pain may be worth it ? Still, while I believe that ship has sailed others disagree that's the case. At a minimum, the other browser vendors are categorical raw font linking is a fact on the ground.

So much for monolithic Microsoft thinking, I guess. If anything, do me the favor to remember that I work for Microsoft. I am not Microsoft. So do engage me. But it's going to be a bit easier for you and me both if we don't have to go through the stereotypical "Microsoft is stupid and evil but of course I don't mean you personally". Now if you do suspect I'm evil and stupid, well, at least ask yourself why I'd be taking the time.

Best,
S.
Received on Friday, 3 July 2009 00:24:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 11 June 2011 00:14:02 GMT