|
About maintenance of MSOOXML should it become an ISO standard... and some ideas - Updated |
|
Thursday, December 06 2007 @ 02:29 PM EST
|
Rob Weir has discovered something so awful about the maintenance of MSOOXML, should it become an ISO standard, I wanted to make sure everyone knows about it, and then I have some projects to tell you about that are trying to respond to Microsoft's gaming of the ISO approval process.
In Weir's article, " Bait and Switch", he informs us that Microsoft is going back on public promises regarding control of the standard, should it be approved, promises which Ecma echoed [PDF]. You can't take your eyes off Microsoft for even a minute, can you?
Here's part of what Weir writes: So much for the promises. What makes this story worthy of a blog post is that we now know that, as these promises were be made to NB's, at that same time Ecma was planning something that contradicted their public assurances. Ecma's "Proposal for a Joint Maintenance Plan" [pdf] outlines quite a different vision for how OOXML will be maintained.
A summary of the proposed terms:
- OOXML remains under Ecma (Microsoft) control under Ecma IPR policy.
-
Ecma TC45 will accept a liaison from JTC1/SC34 who can participate on maintenance activities and only maintenance activities.
-
Similarly, Ecma TC45 documents and email archives will be made available to the liaison (and through him a set of technical experts), but only the documents and emails related to maintenance.
-
No mention of voting rights for the liaison or the experts, so I must assume that normal Ecma rules apply -- only Ecma members can vote.
-
Future revisions of OOXML advance immediately to "Stage" 4" of the ISO process, essentially enshrining the idea that future versions will given fast-track treatment
A critical point to note is that "maintenance" in ISO terms is not the same as what the average software engineer thinks of as "maintenance". The work of producing new features or enhancements is not maintenance. The act of creating OOXML 1.1 or OOXML 2.0 is not maintenance. What is maintenance is the publication of errata documents for OOXML 1.0, a task that must be completed within 3 years. Bait and switch indeed. Now, I'm a simple soul, so I could well be missing something, but it looks to me like at February's meeting, any unresolved technical issues can have Microsoft promising to fix the issue, hence winning approval, after which it, through, Ecma, controls how and when it is deemed "fixed", since voting is by rubber stamp Ecma. Uh oh. Does that sound like a good plan to you? Microsoft controls maintenance of its own standard? That may not lead to true interoperability, methinks. And significantly, some voted for MSOOXML precisely *because* Microsoft promised that if approved, MSOOXML would *not* be under its control. Here's just one of a series of promises made by Microsoft representatives that Weir has collected: For example, John Scholes writes of a Microsoft commitment made at an NCC file format debate held in London in July:
Would the maintenance of the standard be carried out by Ecma (assuming OpenXML became an ISO/IEC standard) or would it be carried out by JTC1? No question, JTC1. But would the detail be delegated to Ecma? No, it would all be beyond MS’ control in JTC1. Well at this point there was apparently some sotto voce discussion between Stephen and Stijn, followed by a little backtracking, but it came across loud and clear in subsequent discussions in the margins that Stephen and Jerry believed this was for real. MS was handing over control of OpenXML to JTC1 (or trying to).
I participated in this debate as well and can confirm, that it occurred exactly as John relates. I even asked a follow-up question to make sure that I didn't misunderstand what Microsoft was saying. They were adamant. ISO would control OOXML. The bottom line is, while Microsoft invents its own proprietary standards process, as I'd call it, it keeps moving forward because everyone else continues to follow the detailed rules and procedures of the ISO process with the laughable results that we have seen with the new P members lined up for Microsoft not bothering to vote on anything else and hence making it hard for committees to do their work. As often seems to happen, everyone must suffer so Microsoft can get what it wants. Convenor Martin Bryan, in his "Working Group Status Report" prepared to report on WG1 activity for the December 2007 Meeting of ISO/IEC JTC1/SC34/WG1 in Kyoto: The influx of P members whose only interest is the fast-tracking of ECMA 376 as ISO 29500 has led to the failure of a number of key ballots. Though P members are required to vote, 50% of our current members, and some 66% of our new members, blatantly ignore this rule despite weekly email reminders and reminders on our website. As ISO require at least 50% of P members to vote before they start to count the votes we have had to reballot standards that should have been passed and completed their publication stages at Kyoto. This delay will mean that these standards will appear on the list of WG1 standards that have not been produced within the time limits set by ISO, despite our best efforts. He writes that ISO is becoming a laughing stock in IT circles and suggests standards that are outstanding be sent over to OASIS. That's one idea. Any others? Since the FOSS side won't game the rules, is there a way to play by the rules and still be effective? Or is it time to change the process to deal with gamers? Weir has a proposal, or more accurately highlights a proposal to set up a new working group in SC34 for "Office Information Languages" maintenance. I am focused more short-term issues, like February's meeting and Ecma's secretive list of allegedly "resolved" comments, by them, natch. If you'd like to work on a project related to that just getting started, I'd like you to either email me or leave a comment on this article, indicating your interest in working on technical comments. Others are noticing how things are going also, and there are a couple of other projects going on that you might like to know about. Norbert Bellow has set up a page for those interested in working on a "problem document" to provide that he hopes the February group will consider: Since it is clear from preliminary discussions that OOXML does not
fulfil OpenISO.org's criteria for an open standard, OpenISO.org will
prepare a "problem report" document explaining the main problems why
from the perspective of OpenISO.org's criteria, OOXML cannot be
accepted as an open standard, and should not be approved as an
"international standard". This "problem report" document should be
ready by mid-February 2008 so that it can help guide the discussions
at the "Ballot Resolution Meeting" to focusing on the most important
questions, and so that after "Ballot Resolution Meeting" the "problem
report" can assist the national standardization organizations in
evaluating whether or not the important issues have been appropriately
resolved. Whether they would consider it, I have no idea. But the public would at least be informed. And you already know about http://www.dis29500.org/. Here's another site that has organized all the comments, as a cross-check. But the project I have in mind and would like to know about your interest in helping with has to do with Ecma's "resolved comments". So please let me know if you'd like to help. And in the interests of long-term educational goals to spur creative thinking, here's a Supreme Court decision about some secretly misusing a standards process to impact the competition negatively, Allied Tube & Conduit Corp. v. Indian Head, Inc., that you might enjoy reading. Here's one bit of the ruling: (c) Unlike the publicity campaign to influence legislation in Noerr, petitioner's activity did not take place in the open political arena, where partisanship is the hallmark of decisionmaking, but took place within the confines of a private standard-setting process. The validity of petitioner's efforts to influence the Code is not established, without more, by petitioner's literal compliance with the Association's rules, for the hope of the Code's procompetitive benefits depends upon the existence of safeguards sufficient to prevent the standard-setting process from being biased by members with economic interests in restraining competition. An association cannot validate the anticompetitive activities of its members simply by adopting rules that fail to provide such safeguards. At least where, as here, an economically interested party exercises decisionmaking authority in formulating a product standard for a private association that comprises market participants, that party enjoys no Noerr immunity from any antitrust liability flowing from the effect the standard has of its own force in the marketplace. The American Antitrust Institute explains what Noerr immunity means, as does The 'Lectric Law Library's Lexicon. If you wish to learn more, here's a book, IP and Antitrust, on Google Books that delves into the subject about as deeply as you are likely to want to go. You'd have to find it in a bookstore or at a library to read it. And some of you might find this paper, RIAA lawsuits and Noerr-Pennington Immunity, by Catherine Gellis, of interest. Here's the abstract: This paper considers whether the voluminous RIAA filesharing litigation would be entitled to Noerr-Pennington Immunity if it otherwise ran afoul of antitrust law, ultimately concluding that, under the Sham Exception to the Noerr-Pennington doctrine, it would not.
Isn't law fascinating? Hopefully some FOSS-friendly lawyers are getting a PhD on the side in the subject. I have no doubt that every move Microsoft makes is vetted by their lawyers first, but as we've learned from the SCO debacle, that doesn't mean they are right.
Update:
Here's more info on the project I am interested in, so you can decide if you are interested too: The New Zealand Open Source Society is
participating in the next round of Standard New Zealand's review of the
Microsoft/ECMA responses to ISO's NB comments for OOXML. You may recall
Standards NZ voted "No - with comments" with regards to the proposed
standard earlier in the year.
Standards NZ will be participating in ISO Ballot Resolution Meeting to
be held in February next year.
At this stage of the ISO process the NZOSS would like to invite any
technically and legally minded people in the free and open source
communities to review ECMA responses: to New Zealand comments or to
comments that might affect New Zealand interests.
We are allowed to mirror responses that fulfill this criteria, and we
have set up a site with this ECMA content for invitees to review and a
mailing list for these people to hold discussions.
They say that many eyes make bugs shallow, so let's apply Open Source
motto that to OOXML. Please email Matthew Cruikshank
(ecma-reviewer at holloway.co.nz) with your full name if you wish to be
involved.
Many thanks for your help
Don Christie
President
NZOSS
|
|
Authored by: racergreg on Thursday, December 06 2007 @ 02:42 PM EST |
Please put the correction in the title. [ Reply to This | # ]
|
|
Authored by: mdarmistead on Thursday, December 06 2007 @ 02:47 PM EST |
As usual, please read the RED info about links, etc. [ Reply to This | # ]
|
|
Authored by: mdarmistead on Thursday, December 06 2007 @ 02:48 PM EST |
Please use the headline in your title [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, December 06 2007 @ 03:17 PM EST |
If you retroactively cancel their status in the last two votes if they fail to
vote twice.
If they fail to vote at all from this, then their P status is revoked.
The first is to undo any gaming (it will be a huge pain, but not as huge as not
having quorum because of stuffing the votes) and the second to make sure that
you don't have to keep checking for slackers in the system (so reducing waste).[ Reply to This | # ]
|
|
Authored by: Nick_UK on Thursday, December 06 2007 @ 04:22 PM EST |
I think anybody that knows MS and anybody that has watched
and seen their actions over the last 15 years expected
this anyway (I certainly did) - hence why the shaking of
heads when ISO was overrun by MS unethical behaviour (and
there is still the question of what Miguel de Icaza is up
too with MS - t[hat will all turn bad too for F/OSS]).
I still don't understand why MS can still get away with
it.
Nick[ Reply to This | # ]
|
|
Authored by: tknarr on Thursday, December 06 2007 @ 04:43 PM EST |
A more interesting question is, are these new P members going to show up in
February? Unless MS throws more bennies their way, I can easily see them
deciding not to go to the expense and trouble. In that case MSOOXML fails
because it doesn't have enough votes (IMO a good outcome).
What ISO ought
to do is implement a "provisional P member" status. When you first apply to be a
P member, you go on provisional status for 1 year. You're counted as a P member
when you submit a vote, but if you don't show up for meetings or submit a vote
on a ballot you aren't counted for purposes of a quorum. In addition, if at the
end of your provisional period you've skipped voting on more than 20% of the
ballots (submitting an "abstain" vote counts as voting) you lose your P member
status and may not reapply for it for 5 years. [ Reply to This | # ]
|
|
Authored by: Alan(UK) on Thursday, December 06 2007 @ 04:55 PM EST |
After the mess you all made of the Off Topic and News Picks threads last time, I
feel justified in starting a thread of my own - definitely no roundabouts.
Starting with OLPC:
Microsoft and Red Hat start from similar positions, both have an OS that runs on
a standard PC. OLPC has made RH Linux run on the X0. Microsoft, having been
given the extra Flash memory required, is struggling to use it because it is not
formated like a hard drive - presumably they expect to see FAT32 and a C:
Microsoft also cannot use the BIOS and needs to rewrite it.
It's a hard life.
Microsoft is trying to install XP despite the fact that it also wants to bury
it. But then they can hardly make Vista run on a desktop.
Life's not fair.
"Finally, we are doing this engineering work for a moving target. It is
literally like designing parts of a car – well, actually a school bus -- while
it is running down the highway at a high speed."
Some people just can't stand the pace.
"Much of the technology in the XO is developed using open source technology
licenses that make it difficult for engineers employed by commercial software
companies like Microsoft to work directly on the project."
Finding it difficult?
Now for MSOOXML:
"This year WG1 have had another major development that has made it almost
impossible to continue with our work within ISO. The influx of P members whose
only interest is the fast-tracking of ECMA 376 as ISO 29500 has led to the
failure of a number of key ballots." - Report on WG1 activity for December
2007 Meeting of ISO/IEC JTC1/SC34/WG1 in Kyoto, Working Group Status Report,
Martin Bryan (Convenor, ISO/IEC JTC1/SC34 WG1).
How to win friends and influence people.
Then - from the Ecma PDF:
"Joint Maintenance Plan for IS 29500 – Office Open XML File Formats"
IS 29500? What's that? Half-way between a DIS and an ISO?
"An Agreement between ISO/IEC JTC 1/SC 34 and Ecma International"
Nobody is agreeing to anything - see above on how to win friends and influence
people.
"ECMA-376, “Office OpenXML File Formats” (referred to below as OpenXML),
was Fast-Tracked to ISO/IEC JTC 1 as DIS 29500."
True, but something is missing - YOU LOST!
I enjoyed that.
---
Microsoft is nailing up its own coffin from the inside.[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, December 06 2007 @ 05:02 PM EST |
In more ways than one this whole affair reminds me of the Internet. I'll
explain:
1) When the Internet was invented (Arpanet actually), all of the procols like
TCP/IP, SMTP, DNS, etc. were devised and implemented in a cooperative
environment. ALL mail exchangers were open relays because this allowed more
chances for a message to get delivered. Once the net became widely available
too all of mankind, it was no longer a cooperative environment. In the present
hostile environment, the old protocols don't do so well. They've all been
extended to deal with various security vulnerabilities but there are still
problems (like spam). The lesson: Systems that are designed to operate in a
cooperative environment will break down when operating in a non-cooperative one.
This applies just as much to ISO commitee bylaws as it does to network
protocols.
2) Now for the other analogy: The Internet was a great place until Microsoft
came along. Most of the problems with the net are due to large armies of spam
zombies. They are all hacked Windows boxes.
JSL[ Reply to This | # ]
|
|
Authored by: CraigAgain on Thursday, December 06 2007 @ 05:28 PM EST |
"indicating your interest in working on technical comments"
I'm very interested and think I can help.
---
Don't change your dreams to fit reality. Change reality to fit your dreams.[ Reply to This | # ]
|
|
Authored by: gdt on Thursday, December 06 2007 @ 05:40 PM EST |
Here's what I wrote in my 2007-07-20 submission to Standards Australia.
Presumably Standards Australia collated this issue and it is now one of the
issues in front of JTC1 -- although it is impossible to tell since the list of
issues to be considered is being kept secret by ISO. Nor has Standards Australia
issued a list of the contradictions they sent to ISO. Unfortunately Standards
Australia is an extra-governmental body, so the usual measures such as Freedom
of Information requests are ineffective. The Memorandum of Understanding between
Standards Australia and the Commonwealth of Australia requires Standards
Australia to act openly and transparently, but there is no enforcement mechanism
should this fail to occur.
Maintenance
authority
If DIS29500 is approved there needs to be adequate thought
given to
the maintenance of two similar standards which exist in the
same
field.
It would not be in the national interest to have distinct
maintenance
authorities for these two standards.
For example, both
specifications need work maintaining spreadsheet
formulas: the specification is
deficient in both specifications and
both have functions which return US-centric
values. By a historical
accident formulas are essentially the same between the
standards, as
both come from the VisaCalc, Lotus 123, Excel evolution of
formulas.
At the moment this allows spreadsheets to be transported
between ODF
and OOXML: all that is needed is the parsing of the XML by an
“import”
routine (although this is complex enough). If two
maintenance
authorities were to develop two differing extensions to
spreadsheet
function formulas then fixing the semantic differences between
the
spreadsheets would be difficult if not impossible. This would lead
to
maintenance making the interoperation of OOXML and ODF even
worse.
The maintenance authority should not be ECMA, you have seen the
poor
quality of their work in DIS29500. OASIS would be acceptable,
particularly
since OASIS' function is the development of document and
work flow
specifications and since Microsoft is already an OASIS
member. Better still, ISO
itself should maintain the two standards so
that the behaviour of all parties
can be closely supervised.
Glen Turner [ Reply to This | # ]
|
|
Authored by: Brian S. on Thursday, December 06 2007 @ 07:06 PM EST |
relies on respect for the body which sets them.
ISO is the closest thing
I know to wordwide standards, BUT ISO is having it's name dragged through the
mud to such an extent that it must give them "food for thought". Why should
China or Vietnam accept ISO's version of an obscure "Americanised computer
standard" even if they do go along with the French Metre or British Greenwich
Mean Time.
ECMA is dead. It's a European body taken over by an American
company. I can't think of a better way to commit suicide.
Brian S. [ Reply to This | # ]
|
|
Authored by: cmc on Thursday, December 06 2007 @ 07:59 PM EST |
Maybe it's because I don't really understand the relationship between ECMA and
ISO, but I'm confused. If MSOOXML becomes an ISO standard, then why should
maintenance be controlled by ECMA? Shouldn't the maintenance be controlled by
the organization which created/adopted the standard (ISO)?[ Reply to This | # ]
|
- I'm confused - Authored by: Anonymous on Thursday, December 06 2007 @ 11:52 PM EST
|
Authored by: Ian Al on Friday, December 07 2007 @ 05:08 AM EST |
I want to be part of the angry mob advancing on Redmond with flaming torches and
pitchforks.
The time for reasoned argument is long past.
---
Regards
Ian Al
Linux: Genuine Advantage[ Reply to This | # ]
|
|
Authored by: attila_the_pun on Friday, December 07 2007 @ 05:41 AM EST |
Those members who persistently fail to follow the rules on voting should have
their votes invalidated.[ Reply to This | # ]
|
|
Authored by: Jeetje on Friday, December 07 2007 @ 05:53 AM EST |
The solution is quite simple imho: any member who has not been P member from the
very moment a standards proposal entered the ISO standardisation process
(whether it is through the 'fast' or the normal track) can only vote as an O
member.
To take that even one step further: for any new standards proposal,
the default status of a member should be no higher than O member. As soon as a
standards proposal enters the process, any member should indicate
immediately whether he wants his status raised to P member for this
specific proposal. Missing a ballot as a P member during the process would mean
immediate demotion to O member without the chance for promotion in the track of
that specific proposal. Voluntary demotion is possible at any time, promotion is
not unless the member has spent more time as a P member than as an O member
in the track for that specific proposal (in order to make it even more
difficult to gamble the system at a later stage of the process by bluffing your
way through it).
These simple rules would force any 'player' to put his
cards on the table very early in the game. They foster the idea of promoting
open discussions amongst the different stakeholders, meanwhile keeping party
crashers at bay.
[ Reply to This | # ]
|
|
Authored by: Anonymous on Friday, December 07 2007 @ 09:31 AM EST |
I am just glad microsoft wasn't around when the calculator was being developed.
they would probably make models that would only add and then we would have buy
additional licenses if wanted to subtract, divide, multiply and then have
patents on it so they would be the only calculator manufacturer out there.
[ Reply to This | # ]
|
|
Authored by: Peter Baker on Friday, December 07 2007 @ 02:45 PM EST |
From a Risk Assessment point of view this is the stuff of
nightmares.
Microsoft has by their actions broken the mechanism to ALL
business interoperability, from the correct dimensions of an M8 bolt to formal
quality control and risk management processes. All that damage, just so it
could hang on to a office software monopoly.
This has gone well past the
point of acceptance, and especially those who wilfully participated in this scam
are in principle liable for the damage their efforts will cause to business
worldwide, and that damage is already starting to show. At the moment it's
mainly technical, but tell me, as a random grab from existing standards ISO
holds, would *you* want to be responsible for, for instance, holding up a
standard for Respiratory tract humidifiers for medical use? Or a standard to reduce
injury to children's heads? I'm not sure if anyone really understands just
how catastrophic the damage is that Microsoft has caused with its shenanigans
but the ramifications are IMHO very worrying.
I have reported the matter to
the office of Dr Neelie Kroes, to be considered in the light of the recent
monopoly judgement against Microsoft. I must admit that I don't quite know what
CAN be done, but this act of wilful destruction must not go unnoticed. If you
know any journalists, PLEASE mention it to them. Show them facts, not guesswork
- there are enough. Get it public.
In my opinion, ISO needs to do some
serious house cleaning. In the core organisation, rules must be strengthened
and an arbitration panel must be installed to ensure an appropriate means is
available to deal with any future abuse. Given the proven abuse of the
mechanism, MSOOXML should be wholly removed from the ISO voting process until
the effects of the vote rigging have been negated. The ficticuous P members
that have joined but not voted must be removed on account of joining on false
premises, their payment considered compensation for the harm they caused.
Repeated non-voting must be set as a reason to lose membership status.
In
the national bodies, several processes are ongoing already to establish how
Microsoft managed to sway the committee leaderships. In countries like
Switzerland formal complaints exist against the leadership, and in general there
has been a global discovery that what was supposed to be a fully technical and
logical process has emerged to be a human process that can be easily subverted
by companies with ill intent. This too needs addressing but is much harder
because the mechanism is not universal, it differs per country. I don't think
ISO can do much about that centrally.
I have been involved in cleaning up
organisational messes before, but I must admit I've never seen anything on such
grand a scale. This is not just unethical, it's wanton destruction.
If
Microsoft has ANY sense whatsoever it should start undoing the damage it has
caused, starting Monday latest. If not I would hope everyone reading this will
do their best to ensure that the full effects of what they have done are made
clear to every press outfit on the planet.
This .really. has gone too
far. --- = P =
[ Reply to This | # ]
|
|
Authored by: darkonc on Friday, December 07 2007 @ 06:09 PM EST |
If the rules require that P members vote on all relevant issues, and groups of P
members have not voted on such issues, could it simply be taken as given that P
members who have missed more than some percentage of their votes (1/2, 1/3) in
the last year or two should be deemed in violation of their mandate?
This
would make it pretty easy to toss one-hit wonders, while allowing long-term
members to miss a vote, now and then (as may sometimes happen with emergencies,
etc.), without being tossed.
In addition, a rule should be added that, to
become a P member, one should first be an I member with a good voting
record (I.E. one that would not have gotten them punted if they had been P
members for the last year or two).
In other words, anybody can be an I
member, but to be a P member requires you to actually have a decent, recent
track record.
It's not perfect, but it makes it a bit harder to game the
system.
If being a P member requires voting, then one would hope that
there is some sort of rule, somewhere, that allows some form of
enforcement. --- Powerful, committed communication. Touching the jewel
within each person and bringing it to life.. [ Reply to This | # ]
|
|
|
|
|