On Tue, Jan 17, 2012 at 12:02 PM, Dennis E. Hamilton
<[email protected]> wrote:
> Rob, I agree completely with your sentiment concerning slavish adherence to a 
> specification when that adherence breaks interoperability in unfortunate and 
> costly (support-wise) ways.
>
> My only point is that such slavish adherence also exists in changes that show 
> up in OOo-dev 3.4.0 and in the current developer snapshots at Apache 
> OpenOffice.  So there is a pot-kettle-black problem in making derisive 
> comments about how LO has also fallen into that pot-hole.  Making 
> attributions about TDF process quality is unwarranted.
>

The Apache project was not responsible for the 3.4 beta.  TDF was
responsible for what they released.  When we release 3.4 we will be
responsible for that.  I sure hope that we don't have that bug.

> It would be good to put our (AOO, LO, etc.) heads together and ensure that 
> there is a mutual agreement on a way to preserve interoperability and also 
> make the deviation from the specification widely understood (i.e. with an ODF 
> Interoperability and Conformance Advisory and, ideally, an ODF 1.2 erratum 
> and an ODF 1.3 correction).
>

It would be good to know what Microsoft will do with their next
version of Office as well, e.g., whether they will have this same
issue.  I'd hate to make a change, and then have next version of
Office suddenly require that attribute and reject docs otherwise.
Maybe something to try at the Brussels Plugfest.

>  - Dennis
>
> -----Original Message-----
> From: Rob Weir [mailto:[email protected]]
> Sent: Tuesday, January 17, 2012 04:27
> To: [email protected]
> Subject: Re: OOo, ODF 1.2 extended format and Word: is the possibility to 
> open OOo generated file by word ed as bug?
>
> On Mon, Jan 16, 2012 at 10:03 PM, Dennis E. Hamilton
> <[email protected]> wrote:
>> Rob,
>>
>> I have no idea how get from a catch-22 between ODF 1.2 (latest), ODF 1.2 
>> earlier, and down-level ODF 1.1 consumers that are strict about it (like 
>> Microsoft Word 2007/2010) to this outburst about LibreOffice.  There is no 
>> foundation for that departure into the stratosphere.
>>
>
> I have no wish for the ODF standard, like the US Constitution or the
> Bible, being used as an excuse to justify stupidity.  ODF is a
> specification for document exchange.  If you are using it in a way
> that decreases interoperability then you really need to step back and
> ask yourself if your literal interpretation really makes sense.
>
> If you bought a hammer and the instructions had an ambiguous statement
> that lead you to think it was asking you to hit yourself in the head
> rather than the nail on the head, would you do it?  What if it was an
> ISO standard?  I assume not.
>
> Think of it from the user's perspective.  Are they really going to be
> satisfied with the argument "The standard told me to do it" when their
> documents show up as corrupt in MS Office 2007?  I don't think so.
> Standards may have errors.  User expectations do not, at least not for
> any viable software application.  And until we have 90% market share
> the user will assume that MS Office is correct.  Reality sucks.  Deal
> with it.
>
> -Rob.
>
>>  - Dennis
>>
>> FOR THOSE PLAYING ALONG AT HOME:
>>
>> True, the situation has been over-simplified and everyone who says not-us 
>> presumes it is some other guys instead.  But that is not what analysis 
>> reveals.
>>
>> For the record, simple documents written in ODF 1.2 (extended) from the 
>> Apache OpenOffice developer build of 3.4.0 have the same down-level interop 
>> problem with Microsoft Office.  A simple ODF 1.2 (not extended) document 
>> from OOo-dev 3.4.0, the last beta produced by Oracle, also has the problem.
>>
>> A simple test against the Microsoft Word ODF 1.1 consumer with *any* 
>> document from *any* of these producers could have detected the problem.
>>
>> However, a simple document written in ODF 1.2 (not-extended) from 
>> OpenOffice.org 3.3.0 (Oracle distribution) does *not* have the down-level 
>> interop problem.  Which means that, instead, it is in violation of the ODF 
>> 1.2 manifest schema, which was changed after anticipatory claiming of ODF 
>> 1.2 happened.  You will be thrilled to learn that Lotus Symphony 3.0.0 FP3 
>> is in the same boat: Word doesn't flinch, but a current ODF Conformance 
>> Checker will.
>>
>> My money is on those who realize that the schema is not the thing to be 
>> slavish about, especially for an inconsequential retro-active provision, if 
>> the goal is least friction for interop with their own down-level and other 
>> ODF 1.0/1.1 implementations.  It is clear that, with the confusion this 
>> situation has caused here and in the interop efforts of actual users, the 
>> creation of this retro-active provision in ODF 1.2 was a stupid move at the 
>> ODF TC.  I am embarrassed that I didn't catch it there.  Basically Michael 
>> Brauer and I are the culprits.
>>
>> My direct experience with senior LibreOffice developers is that they are 
>> acutely aware of changes in ODF 1.2 that are causing needless down-level 
>> interop problems and they are attempting to navigate that morass.  At Apache 
>> OpenOffice the development effort is just getting revivified enough to where 
>> some of these same provisions can be dealt with.  It would be good to find a 
>> common basis for navigating these shoals.
>>
>>
>> -----Original Message-----
>> From: Rob Weir [mailto:[email protected]]
>> Sent: Monday, January 16, 2012 12:23
>> To: [email protected]
>> Subject: Re: OOo, ODF 1.2 extended format and Word: is the possibility to 
>> open OOo generated file by word ed as bug?
>>
>> [ ... ]
>>
>> If a program does not meet user expectations then it is a bug.  If you
>> want to be compatible with Microsoft Office then you need to play by
>> their rules.  The existence of standards like ODF and OOXML does not
>> change the basic fact that interoperability is hard work.  It requires
>> testing.  It does not happen overnight. It is not merely the result of
>> an incantation that begins with the sacred syllables "ISO".
>>
>> In any case Seeing responses like this from LibreOffice makes be very
>> optimistic about the future of Apache OpenOffice.  Whatever the cause,
>> the fact that LibreOffice ships with this problem shows either a
>> woefully inadequate QA program, or total indifference to real world
>> requirements.  Even testing a single LibreOffice document in Office
>> 2007 would have shown this bug.  Is that too much to expect?
>>
>> [ ... ]
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [email protected]
>> For additional commands, e-mail: [email protected]
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to