Tim Bray’s comment on Secretary General Bryden’s comment and a redux of Billy G crushing of Netscape

Time Bray let loose what he thinks will happen with ISO OOXML approval. If I read him correctly, he is saying nothing will happen, i.e., it will be approved. If so, I agree with him. I do not think there will be any last minute upheaval. I will be surprised if there is. Nevertheless, I think the appeals should be lodged, they make it clear that there are National Bodies who cares about the ISO standardization process and forced ISO to take its head out of the sand and render a judgement.

While I agree with Bray that “Let’s bear in mind that a favorable outcome for the appeals would mean, in effect, ISO acknowledging that they’d made a big high-level mistake. And then let’s not hold our breath waiting for transparency or neutrality. “, I will say that since ISO is mainly a Western organization rather than a “Chinese one”, we have better chance of seeing ISO swallow the bitter pill. Even then, I severely doubt it. If ISO wants to nip it, it has plenty of chance after the BRM and before the appeals. If there is one, the necessity for the BRM muddle through 800+ comments is the final nail in the coffin that OOXML is a technical failure.

This week everyone is bombarded with the news that Bill Gate is stepping down as full time Microsoft employee. In this bombardment of news story, we see the US AntiTrust suit against Microsoft for crushing Netscape see the light of day again. A lot of commentators latched on to the fact that what Microsoft did wrongly in that case is the overzealousness in crushing Netscape. Bundling/exclusivity agreements are the normal tools of the trade, nothing wrong with them and they normally increase competiton and benefit consumers. Therefore, under normal circumstances, Microsoft is on the right side of the law. However, there is a very thin line between agressive competition and abuse of monopoly power, which Microsoft step across to the wrong side in the browser war. To tell the truth, this line is so thin that even today I find it difficult, even retrospectively, to see where agressive techniques by Microsoft definitely cross the line in this case. However, cross the line they did. In most of the important, insightful analysis of Bill Gate’s and Microsoft fortunes, the commentators’ agree that Microsoft did not know when the cross the line. Bill Gate’s obvious irritation when giving his deposition more than amplely demonstrated this fact. I believe most of his “I do not recall” is acting prescribed by his lawyers. However, the body language is his. My other take in the deposition is to ask everyone who bought Microsoft Computer Dictionary released in 1997 to ask for a refund, since it cannot even get a definition for common terms that its company supremo Bill Gates can agree with.

Fast forward to OOXML passage through ISO. Microsoft used very agressive tactics. I have also no doubts that other parties used similar tactics before. The differences is, this time, it is perceived that they outdo others in agressiveness and fatally for them, they did it in public, when everyone is watching (and waiting). Again, Microsoft know it is using very agressive tactics, but probably do not think it cross the line between agreesiveness and rule bending.

Did I believe the rules are bended? Yes. they are bended beyond breaking point. ISO must put a stop to it happening in future, and decides what to do with the current OOXML case. As a “face saving” measure and a compromise, I am willing to accept that OOXML as ISO standard (to give Microsoft a way to step down from the podium), but (in upholding the principle that OOXML is defective) will insist that the next revision of OOXML must overcome all the problems identified, no matter how long it takes. If it takes 10 years, so be it.

Why a free java is good for developer

Neil McAllister “Java is Free at last, so what?” article puzzle me. His generalization that before Java was Free, Linux application developers has a “dilemma” on whether to use Java or not for the fear of alienating Free Software supporters in my mind just simply do not exists.

First and foremost, there are a _lot_ of people writing proprietary Linux application. To them, they don’t care whether Java is free since free software users are not their client base. No dilemma there.

Those of us who count Free software supporters inside our customer base knows that Free software supporters come in many shape and kind. There are a portion of ardent free software supporter that will simply not use any other software. If they formed the bulk of one’s target base, not being allowed to use Java is just the norm as they would had accepted that they have a more restricted set of software tools available.  So what dilemma?? To be truely honest, that position is simply just another position. With the proprietary software route, one’s ability to use software depends on one’s ability to pay. This would had in a lot of case restricts the available software for you more severely than free software do. In a lot of ways, restricting oneself to free software is ironically, liberating. The fact that you know you do not have to deal with a large class of potential problems, such as keeping up licensing fees, forced to rework your software is in the long run, a fringe benefit. The best part is actually being able to resolve bugs yourself, rather than convincing another person there is a problem, then wait for them to fix it. Moreover, the more you dive into the Free Software Universe, you will find that you have a lot of quality software to work with to build your own software.

Most Free Software supporters are like me, practical people who wants and will make sacrifices to write free software; When it comes to buying in software, we will take free software over proprietary one if both are equivalent in functionality, and will even give Free Software the edge if free software is slightly more awkward to use and does not have the features that we like but can live without.  Richard Stallman’s Java Trap argument is read simply as a warning that there is risk with Java. Most importantly, RMS did not tell us to stop writing in Java. He just asked us to be selective in the java VM vendors we choose.

Now that we have a truely free Java, what’s the upside? First and foremost, it simplify the software deployment process. And with this, I mean for all developers. Most users of Linux uses a distribution to manage their software, and more likely than not they got their hands on a distribution that only do Free Software on initial installation. This means no Java by default and this complicates the installation process for Java-based Program.  Take me for example, I put a copy of Java VM in every download. It is either this, or I can instruct my users on how to download and get Java.  Most people like me will choose to bundle java: Less hassle in the long run.

And a free Java just extended the horizon of free software available immediately!  Being practical and working on Free Software using Java with others work on freeing Java means now that Java is finally freed, software do not have to play catch up but can compete immediately with other software. Of course I cannot take credit for this. It is those who write the Free Java that earns all the kudos. I also know this is a choice that RMS would not had made, but I am not RMS.

So, Free Java is a major milestone for developers, All develepors.

ISO Secretary General’s comment on OOXML

Via Alex Brown Blog Entry, we now know what ISO’s Secretary General thinks about the ISO OOXML process. I agree with Brown for saying that it is now important for ISO to learn from the process, and will go further by saying that whatever changes tha comes, it should not be retrospectively applied to ISO OOXML. The only way ISO OOXML should be changed now is that it is demonstrated that current procedures (in spirit or in words) are violated and the violation are serious enough that intervention is warranted.

One other thing to note is that Secretary General Alan Bryden is unlikely to say that the Fast Track process is abused. That would be admitting something wrong with a process he is tasked to supervise. It is true that Europeans are more likely than Chinese, i.e., me, to admit mistakes, but there is still something we Chinese call “faces” that is universal whereever you come from. However, most importantly, even if he personally thinks that the process is abused, and there is NOTHING that says he does, he cannot say it as it will be rightly seen to seriously prejudice the current appeal process.

And of course, his job is like that of Alex Brown when he was the convener of the BRM: He must not take side and stick to his job of smoothing everything over, trying to reconcile conflicting view or in other words, do whatever a aribitrator/manager is suppose to do. As such, he is duty bound not to make any statement that will cause resentment from anyone involved in the process, ranging from Microsoft to the National Bodies that appealed.

I could also choose to follow Alex Brown by emphasizing Sec Gen Bryden’s comment on “experience learned … improvement to follow” but I am worried that I would be making a wrong emphasis on his comments. I do, however, hope that this is indeed what Sec Gen Bryden is going to do. All I will say is lets see what ISO does in the near future. Changes in an organization as large and diverse as ISO will no doubt be a very slow moving train. I would not be surprised if the changes originating from this experience only starts to trickle in from the end of next year.

Listing software prices separately is the first step (and perhaps the best compromise)

PC World is running a story about a French court ordering PC vendors to list software prices separately, but short of ordering vendors to sell computers without operating system. UFC-Que Choisir, the French open source group is not happy with it and will appeal to ask a higher court to order the latter.

As a lot of commentators will acknowledge, listing software prices separately is an important step. The most important achievement here is education of the joe consumer, since it makes them more aware of how much they are paying for the individual parts of the computers. To a lesser extents, for those who wants to take advantage of vendor’s pledge to refund money for unwanted software, this makes it easier. This will put an end the practice of PC manufacturers understating the actual cost of software to discourage claims, since doing so will make them fall foul of trade description act. However, the best deterant for this is probably it getting into hot water with the software suppliers since other manufacturers will start to demand the same price.

This is more controversial, but I do think the court is correct in not requiring PC vendors to offer naked PC (PC without OS). I agree with the court that there is value in pre-installed software. As long as the right for refund of unwanted software is protected and enforced by the courts, requiring the sales of naked PC can be argued to be intrusive to business. There is nothing to say that selling naked PC actually benefits people who wants it better. It is possible that the cost of creating a separate line of “naked PC” more than outweight the cost of handling refunds for unwanted operating system and in this case, the consumer loses. Therefore, it is better to leave that to the business to decide. After all, if the cost of handling refunds far exceed the cost of a separate product line, guess what will happens next?

Did Windows “N” edition failed??

Via LinuxWorld, I came across an article in Northwestern University Law Review discussing the Microsoft-SamBA agreement, and via citation 8 of the article, a link to SSRN for an article by William H Page which claims, on page 16-18, that EU’s decision to force Microsoft to provide Windows “N” edition, i.e., without Windows Media Player, failed miserably.

I have nothing to add about the Microsoft-SamBA’s agreement except to say that the authors of the NorthWestern University Law Review Article could had seen the agreement as potentially a “third way” to manage the patent quirkmere we find ourselves in, i.e., a full disclosure of the patents one party says it has and where it will impact for all to scrutinize and challenge rather than the expensive, the-party-with-most-money-wins patent lawsuits. Moreover, in a lot of respect, it is a more balanced agreement than usual, given that it not only define a route that both parties in the agreements (SamBA and Microsoft) can work in harmony on a really torny issue, but also protect third parties with the setting of an expiry date after which no claim that the person used privileged information simply because he saw a document under Non-disclosure agreement can be filed.

I have to say, being an outsider working in the EU, I have to disagree with with the claim in the SSRN article that Microsoft’s “N” edition failed. A few facts that I disagree with the authors. Before I mention them, I have to say that I am not a lawyer and perhaps fact alone makes me interpret things wrongly. First of all, I do not think EC did not say that Microsoft can charge the same for “N” edition. Nor did it says the “N” edition cannot be less functional. By definition, removal of Windows Media Player means “N” edition is less functional. It says that customer cannot be disadvantaged for choosing the “N” edition. This translate to having ALL functionality that Windows must have with exception of the Windows Media Player, and also means that customer cannot be made to pay for Windows Media Player that did not come with the “N” edition. After all, if you have to pay for Windows Media Player which you do not need, you are at a disadvantage by definition (again)! Another theory on “disadvantaging” which I might decide to check with the UK authority to test is whether retailer must sell me Windows Home Basic N at a price consistent with their price for the standard edition because I cannot find Windows Home Basic N edition. I admit my theory, if accepted, is pushing the envelope of “disadvantaging N edition user” since there will be a lot of economic factor at play.

Second, EC says Microsoft must create “N” editions for ALL Microsoft Windows product on sale, not only OEM. To me, Windows “N” edition had not failed. I am one of the few that original believe the “N” edition will not have much impact, until recently where I find that the “N” edition will suit me fine and that I can get it cheaper than Windows with Media player, at least in theory. I am sure a lot of businesses will find this is the case as well since we can get the more expensive versions of Windows “N” edition. Don’t believe me? Go and take a walk down PC World and you will find them. [For those of you in not in Europe just search for Windows “N” edition on any  reasonably-sized UK-based retailer]

Back to the article. I do not know why the author choose to restrict his discussion on the unbundling of Media Player on the OEM business only. While it is a big section of the technology hardware for Microsoft Windows, it is not the only business in town. The increase popluarity of Mac, with its dual boot possibility, and the rise of virtual machines will decrease the dominance of OEM business for Microsoft Windows day-by-day.

It is true that today in the OEM segment of business, it is difficult to find “N” edition of Windows. However, I think it is only a matter of time. When the US Antitrust judgement first came out, it took a while before the first OEM dare to ship a rival operating system preinstalled on computers. Three years ago I would not had thought PC World will be selling “N” editions but they do today. This might just be enough to get the momentum going for us to see “N” editions on computers three years down the line.

Could the authors be possibly trying to setup a strawman when they concentrated on the OEM market alone? It is tempting to criticize them that way, but I think not. Rather, I will say the authors do not have their hands on the pulse of the overall technology business at large. After all, the article demonstrates that their expertise is in law, perhaps with a slant towards technology related law, not the business of pushing hardware/software.

UK Unix User Group legal challenge and did I sense Alex Brown’s frustration?

Alex Brown blog post on ODF Alliance’s “favourite” topic, OOXML, updated us on UK Unix User Group challenge of BSI voting decision in the Law Courts. As expected, they flunked. Also as expected, the failure was not well publicized. What was unexpected is the revelation that it fails to show up when it really matter, i.e., when BSI was voting on the issue, and an extremely strong hint that it might just make the difference.

Another interesting point is at the end of the post. I sense that Brown is frustrated at the appeal process delay the maintenance of OOXML. As a technical person, I can understand that. Here, Brown is probably at the lighter end of this frustration since I am pretty sure we are only looking at a delay and nothing more serious than that. This is seen in the context that we live with the “standard” frustration that one can expect in normal business process where our technical opinion is valued but not implemented because it did not make business sense. Obviously, the biggest frustration is that some management people completely overruled the technical committee in a way that we sensed that they simply disregarded our opinion. Hmm… there are at least several people that claimed they are wronged this way.

AP and copyright infringement

It appears that Associated Press sent several copyright take down notice to The Drudge Retort to take down several posting and user comments that is alleged to have infringed its copyright since they cut-and-paste sections of AP reports. NYTime’s coverage is the most balanced view of the topic. Those take down notices created an anger over the net. It appears that AP has now back down in the sense that it will not take things further but still claim infringement. The NYTime’s blogpost clearly show that AP has a PR disaster at hand and is trying to contain it.

The details of the postings are listed at Cadenhead. IANAL, but my view is that it is inappropriate to take the number of words copied to decide whether is there any infringement. In this case, although the actual word count that was copied was low, one must take into account that the actual number of words is the article is also relatively low. My rule of thumb is to dismiss less than 10% of the article is used, than generally I won’t say it is an infringement. In at least two of the quotes, the 10% rule is more than exceeded. However, as I had said, this is only a rule of thumb and is known to be problematic in the case of short articles. For example, if I write two sentences of 10 words each, it will be unreasonable for me to say you infringed my copyright by creating a quote of 10 words. AP’s articles are rather short here.

Moreover, as the NY Time’s post point out, even the Supreme Court do not believe all words in the article are created equal. This complicates the argument of course and make it a very subjective issue.

To complicate matter further, morally I feel that if you had indicated the source of your quote, then you are afforded a bigger tolerance before an infringement kicks in. As usual, the way you use it counts. If the quote is merely 1% of your posting will turn the table slightly in your favour.

In short, it is a complicated thing. Therefore, as far as I am concerned, I think whether we see a case of infringement here is debatable.

However, I know when things went overboard. One of the consequence of this debate is we see someone, in this case Channel Web, digs up AP’s partnership iCopyright program. I leave it to you to ponder whether  charging for using even as low as 5 words from an article is outrageous.

Jaaksi’s clarification on what he think open source should learn from the handset business (Updated)

Jaaksi from Nokia clarify what he meant when he said Open Source should learn from the mobile phone business. From his clarification, and from my perhaps “polarized” viewpoint, the journalist captured Jaaksi’s view point accurately, i.e., there is not much difference between the two accounts.

PJ’s (of Groklaw) view (on her comment about Jaaksi’s blog entry) that “The more he says, the worse it gets.” has some limited, very limited truth in it.

There is always going to be people of both sides who do not want to learn from each other. Jaaksi is not no one of them, and I do not count myself one of them. In our case, we have the traditional handset business on one side, and open source on the other. Both sides has things to learn from others. However, neither can expect the other to immediately ditch its business model. Both excels in their respective aspect: Nokia in bring mobile communication to the mass, and open source in collaboration to write good, solid software that find wide uses.

Anyone telling the other side to ditch their business model is not asking the other side to learn, but is preaching to the other side. Jaaksi’s latest posting sail very close to preaching, but had not yet cross the line. I can predict and most will agree that a great majority open source people own a mobile phone. A not insignificant number of them, including me, own and prefer a Nokia mobile phone. I can say most people in the open source business understand the handset market dynamics as well as Microsoft’s business model because they are part of it. Moreover, the similarity between the two is that most are actually customers of Nokia and Microsoft, not suppliers.

Presumably Nokia will like to take advantage of open source. Before I start ranting, Nokia has experience in using Linux on its phone. And yes, I would also like to see open source take advantage of Nokia and other mobile platforms. However, I will do this by emphasizing what we have in common, and learn from each other. It is no use asking open source to embrace DRM/handset lock in since this is like asking them to throw away their core belief and akin to telling Nokia to give its handset away for free. However, lets note that those DRM, handset lock in etc are details, something we have to tackle eventually, but at present, noting the particular absence of open source on mobile platform like Nokia’s handset, we can set them aside and tackle them later. Who knows, by the time we tackle it, things might have changed, particularly mindsets. Both should take steps to accommodate each other. From open source side, developing apps for mobile platform as they feel the need to do so and from Nokia, a friendly environment. Something like what Perens says is possible is one of the hundreds of possible first step.

Where open source excel is the ability to disrupt the status quo. For business, it breaks the traditional “competition only” mindset to “cooperative-competition”: cooperate to share workload on a common, mundane, no value added, necessary groundwork which are simply money sink, but compete on where one’ can really differentiate oneself from other offerings, such as better experience.

Where Jaaksi list the business built on open source software, one crucial theme emerges: It is those businesses that move closer to the open source model, not the opposite. If you really want to go into the petty quibbling of who open source developers work for, I will go into and argue that the people who seeded those projects do not work for business when they did, and you simply jump on the bandwagon when the project close to being successful. I don’t really want to do this, as I feel that both sides has everything to lose from this type of argument, and there is really nothing wrong for the behaviour that we see. However, if you force me into this lose-lose situation, I will have to make sure I lose the least.

For the handset business for example, we see Google’s forey into it via the Andriod initiative. This is definitely one way to inject open source into the handset business. I acknowledge that Google can do it because it does not have existing customers/vendors to worry about, and this is extremely likely to be the wrong step for Nokia. I would like to see Nokia innovate in other ways to take advantage of open source.

Nokia, being the leader in the handset market, can do a lot of help open source coming into the handset. Its countribution can be as big as IBM leading open source into enterprise. I ask that Jaaksi ignore those zealots in both camps that is going to criticize him whatever he do, and concentrate on improving Nokia’s relationship with Open Source. Most of us in the open source camp are there to learn. We look forward to Nokia’s innovative idea on open source.

[Update 18 Jun 2008] Perhaps Google, in advertising Android, shows why Jaaksi is interested in “educating” Open Source on the current reality of handset business. Google claims that using Android can save handset manufacturer a whopping 20 percent of the total cost. Even after factor in exaggeration by Google, that is still a significant potential saving which can put traditional proprietary source handset manufacturer in a disadvantage.

Nokia to Linux Developers: “Learn the handset business”

By way of Groklaw’s newspick, I came across this article at ZDNet UK titled “Nokia urges Linux developers to learn business”. Good article highlighting the divides between Open source and the current handset practice. Bad title because the article is about “open source”, not “Linux”. Most importantly, Jaaksi,  Nokia’s vice president of software, who opinion is the focus of the article, clearly demonstrate his understanding of this divide.

Most importantly, he is not demanding that Open Source changed to suit current handset business practice. He is hoping to open a dialog between open source and handset industry needs. Therefore  I would not go as far as PJ goes by simply saying “No” to him. I would encourage him to pursue changes in current business practice to get it more aligned with open source practices since I think it will benefit the industry, and Nokia, in the long run. Somethings, like handset lock and DRM is going be difficult to surmount. But lets work around these first and concentrate on stuff that we both believe in. Market dynamics might just shift in time in our favour.

ISO on why we still did not have ISO OOXML published

ISO release a Press Release confirming that four countries are appealing OOXML approval in ISO. That’s not news, we know it already. This Press Release simply confirm it.

What is interesting, however, is the last paragraph. We have cries that ISO haven’t published OOXML according to its rule. I was not sure what ISO rules are so I had refrained from commenting. After all, “publishing” can mean anything from circulating carbon-copied typed text to selected members (ISO own committee members, National Bodies) that can be created in days to a nicely printed and bounded book that takes months to prepare. ISO have not said it failed its own schedule. It still, in my opinion, have not. What  ISO did is to admit that it has not published the standard. It blames it on the appeal process. It says it cannot publish the standard until the appeal process is completed.

On the surface, that sounds fair enough. The standard is being appealed. If we can avoid “repealing” a standard than we should do it. The confusion created by “repealing” a standard is arguably more costly than delaying its publication. Hence, I accept that as soon as an appeal was officially lodged, ISO will not publish the standard until the appeal is resolved.

However, this explanation is flawed when it comes to explaining why we do not see any ISO OOXML publication yet. Unless ISO has the ability to look into the future, it cannot know that there will be an appeal. It can foresee it, in fact, we all foresee it. But as far as ISO machinery is concerned, it has to work as if there is no appeal until the first appeal was officially lodged to it. After all, if no appeals were lodged, then it will be sad that ISO OOXML would not be a ISO standard because of this silly error on ISO part.  ISO OOXML passage is by far the most politically charged and most controversial approval. For ISO to add unnecessary problems to this mix of volatile ingredient is unacceptable.  In short, the first appeal letter is the “stop work” order. Before that, ISO must continue working according to its own rule, i.e., publish the standard.

So, the crucial date is the day of first appeal. In this case, South African official appeal on 28 May 2008. Unfortunately for ISO, they should had published the standard at least a month ago. Therefore, ISO current explanation does not explain why it is not published for at least a month.

My believe is that there is simply too much changes in the BRM approved OOXML standard that the one month date line is unrealistic, therefore, it was not delivered. What this tell us what we already know, OOXML is not suitable for fast track.