[Wylug-discuss] Why Office Document Standardisation Matters to *Everyone*

[Dave Fisher]

On Fri, Feb 09, 2007 at 04:05:37PM +0000, David Holden wrote:
>  Not been able follow this for time reason, in principle I agree with all the 
> above well articulated points. However a different slant on the issue seems 
> to have been given by Miguel de Icaza
> 
> 
> http://tirania.org/blog/archive/2007/Jan-30.html
> 
> 
> Now since the Novell deal with M$ one may be tempted to discount his views 
> however some of his points on first reading seem reasonable in particular 
> regarding the size and detail of the standard.

Thanks for the link Dave, I'd only (mis)read this piece second-hand
until now.

My first reading of it is that it's a very good critique of knee-jerk
"OO+ODF is good",  "MS-Office+OOXML is bad" reactions, i.e.  his
critique of the critics is fine.  

In fact I find myself rather sympathic to almost all of de Icaza's
criticisms of ODF, and SVG, as standards.

But I think that both de Icaza and his enemies are mistaken in seeing
this as and ODF vs OOXML fight (and those who see it as an OO vs
MSOffice fight are even further off the mark).

To my mind, the real issue which people should be disputing, is the best
route to a decent common standard for document formats, i.e. a partially
strategic/political judgement and not merely a technical one.

The key strategic judgements one needs to make should be based on
empirical evidence about the mutability of OOXML and likelihood of its
becoming a level playing field for competition.

ODF may be tied to OpenOffice.org, but that has not prevented other
projects from implementing it, and quickly/cheaply. All the evidence
suggests that this would not be the case with OOXML (remember ODF had
3rd party commitments to implement it before it became a standard).

Moreover, OO's install base is still so small that significant
improvements to ODF are quite foreseeable and would almost certainly be
picked up by OO and the other implementors.

By contrast, Microsoft are almost certain to resist significant
ammendments to OOXML and even less likely to re-write Office 2007 in
response to such ammendments ... not least, because lots of their
customers would whinge about it.  Furtheremore, it's difficult to see
who would want to improve the OOXML standard if only Microsoft are
using it.  

Having many implementions of the same standard (a level playing field)
seems much more likely to produce innovation and competition. 

The competitive pressure on MS to improve its CSS implementation in IE7,
is an instructive contrast.

Like you, I'm uncomfortable with de Icaza's suggestions for 'the way
forward', but not primarily because of his Novell/Microsoft links.

His track record of playing catch-up, whilst simultaneously
slipstreaming, MS platforms (Gnumeric, Mono, etc.) is not one of
earth-shattering success.  If you were already committed to such a path,
MS's 6,000 pages of documentation about OOXML would definitely look like
a 'good thing'.

I've not reached much of a judgement about whether of not Mono itself
has been a 'good thing' overall. 

On the one hand it has definitely excited some interest from a minority
of developers who work primarily (or exclusively) in the MS ecosystem.

On the other hand, 100% of the MS-developers that I've met (and have
tried Mono) have been scathing in their judgement of it (mostly, to the
effect that its .Net subset is too small and/or poor to be of much use).
I'm in no position to evaluate the veracity of such perceptions.

However, I suspect that de Icaza's judgement sometimes gives more
weight to his hopes, than it does to his experience.

If I get time, I'll try to dig up some of his past predictions about how
long it would take for Mono to bury a range of Microsoft-only
alternatives. Suffice it to say, we are still waiting ... and no longer
with baited breath.

Dave