An Antic Disposition

You are here: Home / Archives for 2007

Archives for 2007

Once More unto the Breach

By Leave a Comment

Stephen Walli has a blog, Once More unto the Breach. He writes mainly about open standards/open source, with a solid business/legal angle. He also has hands-on experience with standards development in the IEEE and ISO with POSIX, and an interesting perspective from his broad experience in the industry, including working with standards and community development issues at Microsoft.

Since his blog’s title, like mine, is from Shakespeare, and he occasionally writes about ODF, I am doubly obliged to give a mention.

Looking at the posts tagged ODF, there is some good stuff. In Vendor-Speak: Microsoft and OOXML, he takes a close reading of some of the recent statements from Microsoft on standards and choice, a long-time confusion of terms that I had previously called a language game. Walli points out:

…Standards happen when a technology space matures to the point that customers are over-served and want choice to encourage competition. Customers complaining about price is the market signal. Competitors know they can collectively chase the incumbent vendor with a standard at this point, if they pick the right level of collective abstraction to standardize. This is how standards work in the marketplace. (I would hope the GM for standards at Microsoft knew this.)

A sympathy play isn’t going to work here. Customers WANT the standard that encourages multiple implementations. True Microsoft support for ODF in their Office product suite would have been listening to customers. Complaining that the marketplace is competitive while shoving your own product specification through a standards forum is naive at best and arrogant in the extreme at worst.

Another good post, is How Microsoft Should Have Played the ODF Standards Game:

The interesting thing is to look back on the number of times a vendor with a single implementation tried to win playing an overlapping standards game. Looking at the UNIX wars I remember three occasions off the top of my head where this was tried over a long period.

  • “tar wars” over archiving formats.
  • The GUI Wars where OSF/Motif and Sun’s GUI toolkit battled it out.
  • The sockets versus streams debate.

In each case, we ended up with two standards being forced upon us. In each case, the dominant technology that won in the marketplace was the one with the most implementations, with the other withering on the vine. Even when both specifications became required for an implementation to claim conformance to the single standard that included them, customers in the market used the specification which was most widely implemented every time.

Microsoft chose the wrong strategy here on multiple levels, betting against customers and the market in general. It may buy a bit of time, but will ultimately cost them more in the long run.

This is a theme that Walli repeats in several of his blog posts — the standard with the most implementations wins.

So a hearty recommendation for Once More unto the Breach, a blog that deserves a slot in your feed aggregator.

Here today, gone tomorrow

By 13 Comments

The Ecma OOXML web site has been updated. The version of the OOXML specification which was submitted to JTC1 is not longer there. Instead we have a new version, generated on February 1st. I have no idea if the content of the new version differs in any substantial way from the older version, but it is clear that the pagination is different. So page number citations, as referenced in this blog and other places (such as the Groklaw analysis) are now incorrect.

(Why don’t I cite using section numbers? Good question. This is because the version of OOXML submitted to JTC1 reused section numbers, so a reference to “section 3.4.2” could be ambiguous.)

A more significant change is that the annexes are now zipped up with the PDF file to which it pertains. So Part 4 is now a zip file with 4 electronic annexes enclosed. This is different than what JTC1 received. I don’t believe that JTC1 NB’s received the electronic annexes at all.

Maybe when Ecma finishes deciding how they want to paginate the thing I’ll go back and update page references in previous posts. But for now, I’ll leave them as-is, which will match the version that JTC1 NB’s have received, but not the version that the public has.

I’ll close by saying that this is a bit odd for open standard, that the version that was submitted to JTC1 is not the same as what is available to the public. You would almost think that someone out there did not want public input on OOXML to be easily consumable by JTC1 NB’s.

Merely a flesh wound?

By 14 Comments

The subject is the 19 contradictions JTC1 member countries submitted at the end of their 30-day contradiction ballot. Andy Updegrove broke the story.

Since that post came out there has been some interesting spin placed on these results, spin that I’m seeing popping up in several places. I’ll give you a few examples, and than explain why these ballot results are more significant than some might have you believe.

An anonymous comment on Groklaw:

This is a normal step in the fasttrack proces. (bypassed by ODF however) This might add one or two months to the proces [sic]. The ballot in ISO requires a 2/3 majority. Only 20% have reacted with contradictions or a question for information and Ecma can probalby [sic] satisfy a lot of those 19 by answering questions, expanding on their information and mayby [sic] amending the proposal.

A blog post from Jerry Fishenden, Microsoft’s National Technology Officer for the UK:

I’m not sure where the media are getting their information, but apparently out of the hundred or more ISO members only nineteen of them filed a response to Ecma 376 (Open XML) by the close of the initial 30 day consultation period.

From Brian Jones, whose excellent blog does worthy service to Microsoft’s perspective on OOXML:

The 1 month contradiction phase of the 6 month fast track process is now complete. It sounds like about 18 of the 100+ countries reviewing the standard came back with comments.

And finally, an eWeek article, quoting Tom Robertson, Microsoft GM of interoperability and Standards:

There are 103 countries that participated in the ISO process, and each country has a national standards body with the authority to act at the ISO on behalf of that country.

Obviously, these are their talking points. To see why this misleading, we first need to have a quick refresher on how votes are counted, according to the JTC1 Directives.

First, remember that this is not ISO per-se. It is JTC1, a Joint Committee between ISO and the IEC. Not all 100+ ISO members are members of JTC1. So it is not relevant to talk about how many ISO members there are in total. Only the JTC1 members were eligible to raise contradictions.

Second, the critical number to look at is the count of Primary Countries of JTC1, the so-called P-Countries. There are only 30 P-Countries in JTC1. You can see them listed here. If you compare the list that Andy Updegrove posted you will see that 16 of the 19 countries on his list are also P-Countries of JTC1. So 16 of the 30 P-Countries raised contradictions. The other three are Observers, or O-Countries.

Why is the P-Countries designation important? During the 5-month ballot, approval of OOXML will require that two-thirds of the voting P-Countries approve, as well as that no more than one-quarter of all votes cast are negative. This requirement for two-thirds approval from P-Countries is what makes them so critical.

Do the math. One-third of 30 P-Countries is 10. Add 1 to get 11, the magic number. If 11 P-Countries vote against OOXML during the 5-month ballot, then OOXML will fail. If some countries abstain, then this magic number goes proportionately down. Since some NB’s have a consensus voting procedure for determining their vote in JTC1, the lack of consensus could lead them to abstain from the 5-month ballot, just as it may have lead some NB’s to abstain from the contradiction ballot. So this magic number will likely be less than 11 because of these abstentions.

So, with 16 P-Countries already expressing concerns about OOXML, Microsoft clearly has an uphill battle. If the vote were held today, OOXML would fail in JTC1. To portray the reception of 19 contradictions, 16 of them from P-Countries, as being an average occurrence, or par for the course, or insignificant, is pure spin and and denies the magnitude of the rebuke OOXML has received.

A Barleywine

By 5 Comments

According to the BJCP style guidelines, an English Barleywine is:

The richest and strongest of the English Ales. A showcase of malty richness and complex, intense flavors. The character of these ales can change significantly over time; both young and old versions should be appreciated for what they are. The malt profile can vary widely; not all examples will have all possible flavors or aromas.

Usually the strongest ale offered by a brewery, and in recent years many commercial examples are now vintage-dated. Normally aged significantly prior to release. Often associated with the winter or holiday season.

I started this batch back in November, with Belgian toasted malts (Dingemans Special B and Biscuit) and Target, Cascade and Fuffgle hops. The starting specific gravity (O.G.) was 1.112, which is one seriously heavy wort.

The previous day I had made a yeast starter, building a Wyeast #1056 American Ale 125ml “smack pack” into a 600ml starter (650 ml water 3/4 cup DME boiled for 15 minutes). For high gravity beers this is essential in order to get the fermentation off to a fast start.

After 2 1/2 weeks, the fermentation slowed enough to rack into a carboy where it sat for another month. Today I finally had a chance to bottle this, yielding 11 liters of barleywine. Final gravity was 1.034 giving an estimated ABV of 10.3%, a potent brew indeed. By way of reference, Budweiser is 5%.

An initial taste indicated that it was nicely balanced and hid the high alcohol levels behind the maltiness with forward hints of licorice, vanilla and plum. I will let it bottle condition for another 6-months or so before trying again. This will be a beer to sip and enjoy for several years.

Note that no licorice, vanilla, or plum was ever added to this beer. It is pure beer, according to the German Reinheitsgebot — nothing but water, malted barley, hops and yeast. The rest is the magic of biochemistry, the enzymes released during the malting of the barley that convert the starches into sugar, the carmelization of these sugars during the roasting of the barley, the alcohols and esters produced by the fermenting of the yeast. Even after the yeast has done its work and settled out, the beer will continue to evolve and change over time. Compare the complexity of a serious, living beer like this to the mass-produced, always-the-same pale lagers that fill the store shelves, and you will never go back.

Declaring Bankruptcy

By 10 Comments

Lawrence Lessig called it email bankruptcy: when you have so many unanswered emails in your inbox that you decide to make a clean start and just admit to yourself, and to those who wrote, that you are not going to respond.

I have a related problem, interesting links I’ve collected and have meaning to blog about. But my links have accumulated far faster than I have been able to write about them. So I am declaring “link bankruptcy”. Here is my fire sale, a set of interesting topics for only pennies on the dollar:

  1. Glyn Moody has the story about how platform dependencies has impacted one notable British institution.
  2. Even more startling results in Korea, as reported in The Cost of Monoculture and the Korean Saga.
  3. It is mainly in Polish, but some in English. More coverage of Open Standards in a new blog from Jacek Łęgiewicz.
  4. In case you missed it the first time around, here is a wonderful essay by Dan Bricklin on “Software that Lasts 200 Years“. It made me think of what ramifications this has for file formats that aspire to longevity as well.
  5. This looks interesting. A free OpenOffice Calc add-in for doing “fuzzy math” in OpenOffice.
  6. Sweave adds ODF support to the open source R statistical analysis and graphing platform.
  7. Docvert, an online REST service for converting Microsoft Word documents into ODF format.
  8. I know someone was asking for this a few months ago — A Microsoft Works import filter for OpenOffice.
  9. Office Migration Planning Manager (OMPM) allows bulk conversions of legacy Office binary documents to OOXML. Does anyone have something similar for ODF? Not just bulk conversion, but detection and reporting of possible conversion problems as well.
  10. The eXtensibility Manifesto has some good schema design advice, including: #3 “Design of a data model focuses on all stakeholders’ requirements for the data.” #6 “Designs or components are not reinvented, but rather are leveraged where possible.”
  11. “[Expert Witness] Alepin…alleged that the company [Microsoft]had subverted developers who used Microsoft’s version of Java ‘thinking they were developing multi-platform applications, but were actually developing Windows-specific applications’ “. From PC Pro News.
  12. The Case For ODF — a recent presentation from OpenOffice Community Manager Louis Suarez-Potts.
  13. “Office 2007 lacks some features of earlier versions of Office, and so it can’t fully support some Office files created in earlier versions. For example, Word 2007 cannot open Word files that contain multiple document versions, a feature supported by Word prior to Word 2007”. Anyone know what else is missing? From Directions on Microsoft.
  14. A few months old — European Cities Do Away with Traffic Signs. Does anyone know how this has turned out?
  15. Dashed Lines and their uses.
  16. David Berlind over at ZDNet: “To me, Ecma is not a standards body. As evidenced by the DVD situation (which is ridiculous if you ask me), it’s little more than a puppet with a pipeline through which vendors can pump their proprietary technologies into the ISO standardization process (avoiding the rigor that should normally be applied to anything up for consideratoin as an ISO standard). As such, the ISO is sort of a joke too.”
  17. “One trouble spot we encountered using Vista’s Explorer metadata organization tools was the lack of support for some of the file types we commonly use. For instance, JPEG files happily take attributes under Vista, but PNG files do not. Along similar lines, Vista would not apply metadata to files we had created in the OpenOffice.org format. And, strangely, our attempts to apply metadata to documents created in OpenOffice.org—in Microsoft Office format—were greeted with an error message.” From eWeek.
  18. What is a standard, according to David Rudin, Microsoft’s official Standards Attorney? “A technical specification that enables interoperability between different products and services and is either 1) intended for widespread industry adoption or 2) has achieved wide spread industry adoption.” This is a nice write-up.