The news yesterday is Microsoft has to pull a tool from Microsoft Store because some preliminary (and cursory) analysis by Within Windows suggests it may be contaminated with GPL-ed code. Slashdot has a field day over this story as usual. While the hot heads are calling for Microsoft to open source the tool in question, cool heads are cautioning that this is NOT confirmation that the tool contained GPL-ed code. It is simply a precaution that reputable vendor takes when there is a potential copyright infringement claim.

What I will like to add is Microsoft has behaved very responsiblely. While Microsoft’s mouth need to be washed in strong detergent, their actions when dealing with license infringement is very reputable. For a company chunking out hundred thousands if not millions line of code every day, occassionally license infringement can occur. Not by design but the probability is against them. The trouble that Microsoft is facing is a result of them not being able to control what their mouth says, so any infringement is picked up and amplified a thousand times then a similar news from another company.

I do not believe Microsoft did this intentionally. For those who is gunning for Microsoft paying a heavy price for this act, even if you can prove that this is intentional, they can continue to dream on. The normal penalty is to stop distributing the infringing code, and may be a monetary compensation for past infringement. In this case it appears an alternative is available, i.e., locating and negotiating a license from the author. The fact that the tool was pulled suggest to me that internal audits had revealed that they did not have a license from the author. As some slashdot posters points out, it is the author who can pursue infringement. He can decide to grant a separate license to Microsoft that will “cure” the GPL infringement.

For those saying that Microsoft shold now open the source code to the Window tool, I must respectfully disagree. I believe this penalty is only appropriate as the ultimate sanction if one can prove that Microsoft did this intentionally. Any unintentional infringement should be dealt with by not distributing the code anymore, and the infringer replacing the code or removing the code from any computer with the infringing code.

We must not treat Microsoft differently from other companies. One of FOSS biggest philosophy is equal in dignity before software. It is our duty as advocate of FOSS to make sure we treat Microsoft just like anyone else.

Rupert Murdoch is simply looking for publicity when he says he ‘may’ want to block Google search. We all know he do not want to. This is because unless he has an extremely incompetent IT department (which he don’t), they will tell him inserting a simple robot.txt will do the trick. Google (and Microsoft and Yahoo and other reputable search engine had been honouring robot.txt file for ages.

We all know he don’t want to block Google. What he wants is Google to index his site, then  pay him a fee for directing traffic to his site. In essence, it is “AdWord” in reverse. Rather than paying Google, he is now demanding a fee. Asking Google to pay for directing traffic to news site is not something new and it is a known sore point between search engines and content provider, in particular news provider. It has been bubbling for sometime now. This finally come to a head since Mr Murdoch wants to make his news site a pay-for service.

I don’t care about the argument that Murdoch is trying to build a walled garden. My philosophy is if someone do not want me to view his  content, even if it is free, I will avoid his site. Moreover, it is a capitalist world. He is free to do that.

Precisely because it is a capitalist world, I am issuing a challenge to Mr Murdoch: Take the plunge, setup your robot.txt file to block Google. If your wall garden is so good that Google cannot do without, Google will come to the negotiating table to hammer out a deal. At that time, you can name your price. A more likely situation is Google feels that it can benefit from linking to your site. If so, then we have a negotiation.

Let me tell you what I think is most likely to happen: Murdoch’s site become less popular and probably fall out of the sky… and he cannot bear it.

Someone, I think it is Murdoch, says the only way to break search engine dominant in news stories is for all news site to stop supplying news stories for search engine to index. The person who says this is correct. That is one of the best strategy to bring search engines into the negotiating table. Reversing a trend for free news, I am seeing more sites that start to experiment with charging.

The charging model can work. See Malaysiakini for a good example. They are doing reasonably well because they deliver something others cannot. Can Murdoch do this? I am sure he can. Let the market determines whether it is worthwhile.

Rupert Murdoch’s ploy to get Google to pay him  is like asking Tourist Information to pay an attraction for directing a tourist to them. Quite why the attraction wants to is not my concern. However, let’s not forget, the Tourist Information can choose not to direct any tourist to them as well.

As for his interpretation of “Fair Use” doctrine and that Google breach it? I think it is time for a law court to clarify it. I will love it if Murdoch mount a challenge.

We programmers know about “Many eyes make bugs shallow”, now we have a demonstration that “Many eyes means you cannot bury needle in haysack”. It is about Microsoft trying to bury Novell under a pile of data, as described by Groklaw. This is an age-old tactic relying on the fact that the receipient has not enough man power to sift through the information you bury them under. Of course, the law courts frawn on this type of tactic, as the Novell’s filling will demonstrated. Unfortunately, Microsoft’s argument also demonstrated that determining whether someone is employing this tactic is difficult.

The best solution is of course to put in the resource necessary to uncover the fact. Frankly speaking lawyers employing this tactic know that the art is to build a big enough haystack to bury your opponent in. Too small? They will find the needle. Too large? Beside the obvious unnecessary cost of preparing the larger-than-necessary haystack, it makes it more likely the court will rule that you are deploying this tactic.

In the case of Novell vs Microsoft, the fact that the court documents they are relying on is available to the public and that there are a lot of people interested in it means the resource available to find the needle is much larger than Microsoft lawyers’ expects. And this unpaid army found one. Kudo goes to BoycottNovell for finding one, and Groklaw visitors for finding a few more. Ironic isn’t it? A website that dedicate itself against Novell actually came to its aid.

“Many eyes means you cannot bury needle in haysack” indeed. I do hope this trend continues. I do not think it is right to hide one’s wrongdoing and anything that diminishes the usefulness of this technique is welcomed.

From Groklaw, it is now clear that Novell is prepare to appeal to the Supreme Court. That surprised me. I had never thought that the issue will go all the way up to the Supreme Court.

If I were Novell, I am not really sure I would had tried. It is certainly an avenue available to Novell (and SCO). I would not be surprised if SCO did it, but Novell? According to PJ they must had calculated that they stand some chance of winning. I am going to trust her on that.

Is Novell trying to blead SCO dry? May be. It might also be a negotiating tactic. It wants the Chapter 11 trustee to know that Novell will go all the way and strengthen Novell’s position in any negotiation. Perhaps it is trying to encourage the Trustee to throw in the towel by (1) showing that Novell is not prepared to back down and will do anything to diminish the already extremely thin chance of succeeding and (2) in the process, making it more costly for SCO.

Quite frankly. I don’t know what Novell is thinking. The outcome for the majority of us is an education on US justice system all the way to the Supreme Court by PJ. I had learned a lot from reading Groklaw, to the extent that I know the US system better than the UK system (where I live) or Malaysian system (where I was born). However, I did not expect a complete education, i.e., all the way to the Supreme Court. That is the icing on the cake!

The third part of Rob Weir’s farewell article on OOXML (part 1 and part 2 here) hit a raw nerve at GrokLaw. The argument is in four parts: (1) That Microsoft dominates the working group for OOXML, (2) changes against the consensus at the BRM are pushed through, (3) ISO’s rulse are abused again and (4) OOXML might be reformatted to suit MS Office, not the other way round..

To the third part my reply is “What do you expects?”. There are so many holes in the standard that should not be there. They should had been ironed out before the standard was approved. Unfortunately it is not done so we are now facing the real task of ironing this out. The standard ISO process for standard maintenance cannot cope with it. IF everyone is going to follow the rules to the letter, we cannot finish the task of polishing it up to standard in reasonable time. Therefore, one must push the rules to its limit. I do not like it, but what can we do? For the sake of getting the standard to something resembling the norm, I think I am prepare to accept that the rules had to be bend (or broken) for defects/issues that are not controversial.

This, however, does not mean I am prepare to accept changes that go against the consensus at the BRM (the second point). What the OOXML team promised and blogged about should be kept. They were part of the bargain. I will not accept changing things through the back door normally, so to speak. However, at times, if the consensus at the BRM is altered back to its orignal state, as long as it is reflection of the the current Microsoft documents, I will have to thow my hands up in the air and say while I do not like it, I am not going to oppose it, especially since the charter says it wants to stay true to the current OOXML document.

As to Microsoft dominating the process, if Rob Weir don’t like it, may be he should advise IBM to join it. (In case of any miscommunication, I am saying this in a tongue-in-cheek manner). Let’s see who is in the committee. Pro ODF people, who are the majority of people that has experience in office application? Never there. Apple? Missing in action since OOXML was approved as ISO standard. This is a shame actually. Looking at Apple’s work and  innovation in iWorks, they would have the expertise, experience and innovative ideas to share on the table should they choose to participate. British Library and Oracle? No where to be seen today and like Apple, I do not think they are interested in the continued evolution of OOXML as a standard. So who’s left? Mainly downstream vendors who wants to harness the Microsoft Office eco-system. No pun intended. What I am trying to say here is they do not have the expertise in Office applications and have to rely on others to supply the expertise. And who is there? Only Microsoft.

With only one vendor that creates office application, and that the vendor is a dominant player in the office application field, why should one be surprised that Microsoft is over-dominant in the process? Is there any point in passing, say with only Microsoft decenting, something that Microsoft do not want to do? With no competing implementation to arm-twist Microsoft, that is a stupid thing to do.

Over-dominance by one party leads to undesirable things, such as a standard is seen to be twisted to suit the dominant player. If I were the dominant player and already written the next generation of OOXML, I will of course try to push the standard into the way I like. After all, it is easier and cheaper to change the standard, compared to changing my program code.

If Rob Weir is correct, it will be interesting to find out, if we could, why Microsoft decided to put the brakes on XPS. I hope it is the result of ISO saying enough is enough, we bend over backwards over the abuse of our process (coz we are not prepared for your assault and although what you did is immoral and unethical, it is possible that it is “within the rules”) but we will have no more.

I was surprised to see RMS add his name to an open letter to EU regarding MySQL. Ed Brunette of ZDNet says that RMS finally admitted that GPL is flawed, on the gounds that it is difficult to commercialize software based on GPL. That is certainly one possible interpretation and I can see his reasoning. Equally I can see that RMS will says that he did not say GPL is flawed because GPL, from day one, gave the copyright holders the rights to license their work in other ways and MySQL is exercising this right.

If I am going to call RMS names, I will call him a leech. The reason he wants MySQL spinned off to a third party is because the GPL-ed version of MySQL cannot benefit from the other, proprietarily licensed MySQL. I am not going to call him that because I believe the relationship is symbiotic, with both licenses benefitting from the other. With GPL-ed MySQL, it benefited from development made possible by commercial support for the proprietary-licensed version. However, the proprietary-licensed version benefits from GPL-ed version because the cost of developing the software dropped as they do not have to pay simply to develop their software. Only when the software is ready, they pay per-distribution to keep their software code proprietary.

RMS, in this case, is hoping to keep the symbosis alive and he do not think Oracle stewardship of MySQL is a good idea, therefore, he is camapaigning against it. I don’t really think he had admited that GPL is flawed. The difficulty in commercializing GPL-ed software w.r.t. proprietarily-licensed software is known and I presume, he had acknowledged him long ago.

Steven J. Vaughan-Nichols’ post on ITWorld tells us what we already know: If possible, do not use Windows where you have to transmit sensitive information such as your credit card detail and bank account details.

For on-line shopping. The answer today is a resounding YES. The strangle-grip that Windows (or more precisely IE) has on website is no longer. Nowadays, thanks to Firefox’s popularity, the site will works with whatever browser/operating system combination as they are moving away from IE-centric technology to open standards.

However, I questions the practicality of doing it for banking. Primarily as a result of the high stake involved. Imagine someone fraudulently take a lot of money away from you and the bank says it is from your online banking. Now, the bank is going to scrutinize your computer and say you are at fault because you are running on Linux/MacOSX and you are on your own, regardless of whether the other operating systems are more secure. More interestingly, if you are in UK, the Banking Code (PDF) itself requires you to have up-to-date anti-virus, anti-virus and firewall software installed (clause 12.9). Now, do you really want arguing with your bank that Linux is more secure and do not need all those rubbish softwares? Let’s phrase it differently: Do you really want to spend the money and time to proof that Linux is more secure and do not need all those rubbish in the court of law, if it comes to it? In other words, I am saying you can be better off using an insecure Windows  (with all the rubbish softwares up-to-date) then on a secure Linux or Mac.

Before you say it, yes, we have the same with on-line shopping. However, the credit card company is more mature on this front and is more likely to refund you. Moreover, fierce competition with other credit cards, online shopping facilities such as PayPal and Google Check out means they are more likely to listen to you. For banking we are still in the early stages where the way to deal with problems like this is still being ironed out. This is not helped by the fact that the banks take the position that it must be your fault. Period; End of discussion.

It is a sad end for Darl McBride. He was made redundant due to the bad economy climate. No, not the credit crunch, but SCO’s long-running sorry state of affair. He made a big gamble that by kicking up so much fuss, SCO is bought by someone, and he was targetting IBM. That gamble did not pay off and the chief architect of the scheme has to go. I wish him well on his next job.

More importantly, the Chapter 11 Trustee, appears to be considering keeping the litigation alive. As PJ had pointed out, he might be thinking that the SCO-Novell case might yield something in SCO favour, and if there is something tangile, just may be SCO will be bought out by Novell or IBM, and that will indeed be one of the best outcome for SCO. It is all on a wing and a prayer. I cannot help noticed that that is McBride’s gamble and if that happens, McBride will be vindicated.

The Trustee’s role is to act on the best interest of SCO if possible, and if not, try to bring about as much value as possible for the creditors. Right now the litigation is the best hope for SCO, so it is not surprising that the Trustee is prepared to take the gamble, if financially SCO can drag it out. If this is his intention, we will have to wait for the results of SCO vs Novell to come to fruitation.

Yes, I know. This means the case gets dragged on. However, that’s life.

… more precisely ease of use.

This is the post about what I think after I read the article with the provocative title “Acer: demand for Linux-only netbook not there“. However, it is not Linux that is the problem. Acer is actually trying out Android Operating System, which is based on Linux.

Why Android? Read carefully and you can see the argument is that Android comes with applications, what the Acer guy calls “connectivity”.

Acer’s problem is it is not a software developer. Its competency is shifting hardware. It’s business model is to use software developed elsewhere. There is a lot of software work one have to do, such as selecting the appropriate softwares, test them etc,  if one want to build a netbook using vanilla Linux. Acer would rather outsource it. Right now, Android fills that niche. By pushing functionalities away from the netbook into Google’s web-based services, i.e., the so-called “connectivity”, the hardware requirements drops dramatically, meaning cost is push to as low as a netbook can bare.

However, I am sure if someone can come out with a Linux distribution that has the correct mix of hardware/software/price combo, Acer will likely try it out. Heard that Ubuntu??

PJ believes that Apple and Psystar litigation is developing into an attack against GPL using the “First Sales Doctrine”. May be, but I think it is not. Why? First, the law is there to reflect reality, in real life I see parallels in at least real estate, and I think I believe how the judges will formulate their decision to reflect norm.

In real estate, it is not unusual to have covenent in land title. Take Cambridge UK for example. If you live in or near it, your house is likely to be on ex-Cambridge University land, and if so, your land title  is likely to have a covenent that sounds like this: “The University sells the above-mentioned land to the buyer on the covenent that the land will not be used for commercial purpose. The buyer will observes this covenent and ensure the next buyer in the subsequent sales accepts and observes this covenent and carry on the buyer’s responsibility under this covenent”. In short, the covenent is passed on from one buyer to another. Whether you are the first or 100th buyer, you cannot turn your house into a shop.

Now, there is no question that if you buy a house in Cambridge, you are the owner of the house. Equally, there is no question that the covenent will apply to you. Your act of purchasing the land automatically means you accepted the covenent. The covenent is in effect, equal to a license, or more precisely, a restriction on what you can and cannot do with the land. In this case, you cannot turn it into a shop (or factory).

Second example. [I might have the roles reversed but it is essentially the same story] During the last aviation slump, Airbus provides incentive to airlines to dump older plane by offering the option to trade in their exisitng aircraft, including Boeing’s. This made Airbuses the owner of quite a few Boeing’s aircraft. The sales is real. Although it sounds and taste strange, Boeing cannot say no, because of the first sales docrine. The irony is, being the owner of Boeing’s aircraft, AirBus is entitled to technical updates for the aircraft types they bought. In a sense, you can say AirBus gets intellegence about Boeing’s plane by getting around any NDA Boeing makes their customer sign. Ingenious isn’t it? However, although an owner of Boeing-made aircraft, Airbus still have to obeys Boeing’s rule on what an aircraft owner can or canot do. In particular, it must follows Boeing’s maintenance procedure to the letter, even if it thinks it can do a better job, or disagree with Boeing on the work to be carried out. In other words, there is a limit on what AirBus can do!

Therefore, although I am not a lawyer, I am certain that even if the “first sales docrine” applies to software, the GPL will still applies. In this case it is my view that the GPL, or any other licensing condition your software vendor impose on you, is equvalent to the covenent on the land title, or AirBus having to follow Boeing’s instruction on aircraft maintenance.

Now, how do I square this view? Easy. Being the owner gives you certain rights, but it does not mean you can do whatever you fancy. Owning the house gives you the rights to use the house and the rights to sell it to another party. However, there are still rules that your seller can impose on you and your buyer, i.e. the covenent. AirBus has the right to operate the Boeing aircraft they bought, to sell it when they are done with them, and the rights to access technical documents, but they must follow Boeing’s maintenance regime. Similarly with software: You buy and own the rights to use the software and to sell the rights to use it. Unfortunately, you still have to observe rules that says you cannot reverse-engineer or the rule that says you have to give any modification you made to your buyer.

I will not be surprise if the judge comes up with something similar. Of course, being more intelligent than me, may be they know of a better solution. What I think is really at stake is the software vendor’s insistance that you are a “licensee” and not “owner” and therefore you do not have the rights to pass on/sell the software you no longer need or want. There is a strong likelihood that the terms in your “license” that restricts your rights as owners will be stripped. However, other portions of the license still apply.

As for PJ’s assertion that this is an attack by closed source software against open source. I am not sure. I can see that she based her opinion on the fact that part of the GPL, the part that is different from closesource software will be ruled illegal. It is possible but I do not think this is likely. The judgement, if and when handed down, is likely to equally affect close source and open source software.