CyberTech Rambler

April 30, 2007

Linux installation compared to Windows/MacOSX

Filed under: Uncategorized — ctrambler @ 11:41 am

[Update 20070501: As if on cue, I encounter this article from Information Week describing the installation comparison between Windows and Ubuntu. My take on the article is Windows have the edge, especially because I placed installing hardware drivers as more important than software installation/availability. After all, we are talking about installation, i.e., getting the basic operating system working and what is the most important task here? You got it, getting the hardware "working". I made this decision after taking into account that the computer in question were designed with Windows in mind.

Ubuntu guys need not be disappointed. In fact, I congratulate them for putting Linux on pile with Windows. The hardware driver issues which give Windows the edge is not something in their control.]

As usual, Brian Profitt delivers in its weekly “Editor’s Note” column at LinuxToday.com. This time, he is arguing about the fact that Linux is still not a “work out of the box” solution. He admits, and I agrees, that Linux is still not such a solution yet. However, one point I will like to add is that Windows in particular and MacOSX in general are not such a solution either. The difference is, you get somebody to do the work for you for those operating systems.

Most people, and I am sure it include Profitt, install Linux themselves. This is definitely a different scenario from buying a Windows/MacOSX box directly from vendors. Furthermore, I am sure if you repeat the same procedure for Windows and/or MacOSX, you will have the same problems. If you have the money, go buy Vista and install it on box-standard PC you buy from any vendor. Chances are, even if it has a “Vista Capable” label, you will have problem.

Therefore, the question about “work out of the box” refers to whether you have vendors that are willing to take the pain of dealing with installation on your behalf, not whether any OS “work out of the box”. Quite simply, there is no OS that offers this ability. Probably never. And in this case, yes, we do not have significant vendors that are willing to provide a “work out of the box” Linux…. yet.

A few years ago, the problems with installing Linux yourself is much BIGGER than if you use Windows/MacOSX. However, distributions, particularly Ubuntu, had narrowed the gap considerably. Today I will say the problems with Linux installation is still bigger than the polished Windows/MacOSX but no longer warrant the use of capital “bigger”. Before I forget, kudos to Windows/MacOSX for taking the time to polish their product.

Document Format Usage

Filed under: Uncategorized — ctrambler @ 11:07 am

This interesting article by Nathan Weinberg lists the document format usage that Google Search encounters. Not surprising, Microsoft documents outnumbered Google 477 to 1. However, OpenXML format accounts only for a very tiny and insignificant number. Well, OpenXML format is relatively new, only 3 months. The figure quoted  is in the low 100. I expects it to be low, but am not expecting it to be only 125-ish.  However, it means most of the documents created are in corporate space which are not visible in internet space.

And, … I love the picture in the article. Without giving the plot away, Google search is indeed not almighty.

April 26, 2007

How do you reconcile the fact that Microsoft is using Open Source Software, but at the same time seen to be anti Open Source?

Filed under: Uncategorized — ctrambler @ 3:46 pm

If you ever find yourself needing an answer to this question, think William Hurley. Specifically, one of his reasons in “Seven reasons Microsoft Love Open Source“, i.e.,

They aren’t threatened by open source.

Open source is not the threat; Linux is.  Don’t confuse the two. Open source is growing rapidly, but Linux has several distinguishing features that make it the real challenger.  It’s more mature than other projects, it has a larger, more organized developer base, and it’s well financed.  IBM has spent hundreds of millions of dollars developing, distributing, and advertising Linux, not open source.  Microsoft doesn’t fear open source; it fears what the competition can do with it.”

He rightly says that competition from competitors armed with Linux is the threat but wrong to imply that it is the only threat. Lately, one can argue Microsoft is feeling the heat from ODF. If you think the threat is from IBM use of ODF in its product you will be wrong. After all, how many offices uses IBM software? The threat comes from OpenOffice.org/KOffice/Abiword and other open source office application software which can use ODF to erode Microsoft’s Office dominance. Before this, it is the General Public License “cancer” that Microsoft is trying to eradicate.

The essence of Hurley’s argument is correct, that open source software is a threat only if it is a credible competition against Microsoft. Considering the fact that open source universe is much larger than the sum of all open source software in competition against Microsoft, Microsoft do love open source. It uses BSD software. Based on commentators analysis on a purported leaked version of Windows 98, it appears that Microsoft is itself using GNU Make, the very cancer that Microsoft is once trying to eradicate.

Science Magazine not accepting Microsoft 2007 format and what does compatibility means? [Updated 2 May 2007]

Filed under: Uncategorized — ctrambler @ 12:57 pm

Rob Weir latches on the fact that both Science Magazine and Nature Magazine advise contributors that it cannot accept Microsoft 2007, i.e., .docx format. The reason given is the incompatibility with their internal publishing system. Given that .docx format is relatively new, i.e., less than 6 months, I am not very surprised. It takes sometime for both magazine to upgrade its system to accept new .docx format. The question is will it upgrade in this way and when. At the mean time, they are exploiting their clout to tell their contributors to conform to their working practice, a luxury not enjoyed by journals of lesser caliber.

The big surprise is, that Science advise that if your submission comes from a “downgraded” .docx to .doc format (older Microsoft format), and the original .docx file contains equation, than your .doc file will actually save the equation as “graphics”. Science and Nature are not big equations users. Hence, if the equation problem are big enough nuisance for them to speak out against, how crippling will this be for other journals that uses equations extremely heavily. I am not talking about maths journals which are the expected victims, but computer science journals and engineering journals as well.

This sentence by Science should be noted by everyone claiming that OpenXML’s refusal to use existing standards is not a problem:

“[T]he default equation editor packaged with Word 2007 — for reasons that, quite frankly, utterly baffle us — was not designed to be compatible with MathML” [emphasis mine]

It is powerful because it comes from a disinterested party in the ODF/OpenXML debate. It is powerful because Science technical services and their vendors know that bidirectional MathML to OpenXML Math’s markup is available (See this Brian Jone’s blog entry). It is powerful because it means IT people sees this as a barrier that need not be there. Finally, it does poke a small tiny holes into the claim that MathML is not good enough for Microsoft Office.

Rob Weir claims that this violate the “100% compatibility” claim. This will be true on the condition that we are talking about bi-directional compatibility. As far as I can recall, nobody in the OpenXML camp actually define the term “compatibility”. I so far had taken it for granted that they meant “forward compatibility only”, i.e., from .doc to .docx. Normally, in software one do not expect all information in newer format to be saved in the old format, that is why I only considered “forward compatibility”. If someone on the OpenXML camp wish to clarify that will be great, but I am not holding my breath. [Update 2 May 2007: Brian Jone confirms that OpenXML camp meant "forward compatibility".  (See comment below). Thank you Brian]

April 25, 2007

More on OpenXML vs ODF

Filed under: Uncategorized — ctrambler @ 11:59 am

Wow, did not expect so much pingbacks/comments on my previous post. For a nobody like me, its like Alice in Wonderland.

First, James Governor commented and say that he cannot see 100% ODF in enterprise to the exclusion of OpenXML. As I had replied, its true. I cannot see that either. A few years back I would not believe it is possible to see non 100% Microsoft format, no matter what it is. However, today this is possible. In the next few years, bar some catastrophic events, we will probably see ODF slowly going into enterprise. Best case scenario is it follows the trend of Firefox but slightly faster. Lets say 15% in two years. That is good enough to break the document format monopoly held by Microsoft.

In “Arguing About Archiving“, lnxwalt give a very good account the technical side of consuming a document. It shows that the state-of-the-art in opening MS binary document is less than perfect and implies that it can never be, even with OpenXML, which I agree. Having tried to open a OpenXML spreadsheet document myself, I can truly says I appreciate this quote from him: “It is not just about vendor X supports it, but the field IT person who will have to work with it twenty years from now after the vendors have found something else to attract their attention.”

I can understand why he disagree with me that in developed country like US, governments are more likely to roll over and pay the perpetual tax than to follow my proposal of wading the magic wand that makes the tax goes away. My argument is that IP laws are just legal instruments. A public outcry, the necessity to allow government to continue functioning and the weight of public interest can just conceivably turn those “Men in black” into the alien enemies they are hunting.

What lnxwalt pick up from Brian Jones’ blog entry which I did not was Jones’ claim that “OpenXML allows anyone to build tools that read and write the formats, and at the same time is designed to cause the least amount of disruption possible. You can move all your existing documents into OpenXML, and you won’t lose a thing.” [emphasis mine.] The least amount of disruption to whom? So far, we can see the least amount of work to Microsoft Office Development Team but more work to anyone who wants to read/write it. As highlighted in National Bodies comments to ISO, we have the date problem, non-standard names, the insistence on using alternatives to well established standards such as the mark up language for Maths and finally the opaqueness of Windows Metafile makes more work for others. These are simply design decisions and in my opinion, controversial at best and lousy at worst.

The situation today is that both ODF and OpenXML wants to be your document format. Both took different routes. For ODF, they stress that you can have multiple applications to open it. With OpenXML, Microsoft can hardly point to an alternative implementation of OpenXML available today. Instead, it focus on third parties who are, we should emphasis, not in the Office productivity business but sees value in importing/exporting data to Office applications. In other words, ODF folks focus on competition within Office Apps, OpenXML folks are focusing on building applications around Microsoft Office, i.e., the value-added market. In the long run, ODF is deploying a better strategy. Why? People who build applications around Microsoft Office can easily retarget to ODF, if ODF gets traction. If so, the advantage enjoyed by OpenXML will be eroded. Its an epic battle, with OpenXML folks hoping to put enough effort in the short term strategy sufficient enough to torpedo ODF’s longer term strategy. It just might work.

April 23, 2007

James Governor has a point, but so does Rob Weir

Filed under: Uncategorized — ctrambler @ 12:02 pm

James Governor disagree with Rob Weir over the issues of long term archiving potential of ODF and OpenXML. He has a point there, but so does Rob Weir.

Its true that as long as you have all the bits that constitute your document, you can theoretically retrieve all information in it, even if it is a very difficult task in practice without the document creator’s help. True, it is possible to reverse engineer the format. As Governor points out, Google did it, so did OpenOffice.org. Its difficult and fill with legal landmine that would not be there with a free and open format. Can any single company use legal instruments such as IP laws to stop governments from viewing documents in the future? No. Government has the right to take this legal instruments away. Don’t believe me? see what happen to AIDS drugs in developing country. Moreover, the backslash generated by such a move will probably bury any company who dares. However, as in AIDS drug, the most likely outcome is government or private company has to pay a small sum to the document creator to view the document. Some call this a tax and believe this is unacceptable.

This is precisely Rob Weir’s rub. Should you commit yourself to perpetually paying a third party to view your document? If it is on paper, you don’t. If we buy the argument that saving your document in their format means that the document creator co-own your document, than we must accept that we have to pay the company that produce the ink or the company that produce the paper you write your notes every time you read your note as they co-own your paper document. This will be absurd.

ODF had already brought us one success: Open XML is going through the standardization process because of it. Will it bring more benefits? I think so. It might introduce competition back to the document market. That can be only good for consumers.

Dell selling XP again is not a Vista Failure. Its a hiccup,… and wrong policy to exclude XP

Filed under: Uncategorized — ctrambler @ 11:51 am

A lot of people latching on Dell selling XP again as a sign of Vista failure. I wonder how many times I have to shout this. Vista is not a failure. Its not a spectacular success. At present some will argue it is not yet successful and I take their point. However, it is not a failure. This problem is just a hiccup in the road to Vista adoption.

First of all, it is wrong policy to sell only Vista computer. It looks like a Microsoft’s policy, since Dell is not the only retailer that shun XP in favour of Vista. Sony’s own website did the same with “new models” introduced in January. Those new models are nothing but simply minor hardware upgrade + Vista. True to Sony, you will find it hard-press to find a 6 month old model there. However, resellers do have literally thousands of Sony Notebook with Windows XP on them. As we all know, you cannot keep stock of computing hardware around as they deprecate too quickly. Hence, one conclusion that we can draw is that Sony is still continue selling WinXP, even if it officially don’t. Smart Policy.

If indeed this is Microsoft’s policy and Dell did not complain to Microsoft in the background about this policy, than Dell is more stupid than I thought. Forcing Vista down the throat of consumers at this early stage is forcing “early adopter’s blue” down them and not everyone can handle it. I am glad the cool mind in Redmond or Dell prevailed. This is a wrong policy from day one. The people who formulate this policy is out-of-touch with the market, and in case of MS, we must consider the accusation that the policy makers are just arrogant lot. People working for them who did not voice their concerns are just not doing their jobs. All these leads to this debacle which damage Vista’s image more.

April 20, 2007

$3 Microsoft developing deal fall foul of anti-dumping laws and and OLPC

Filed under: Uncategorized — ctrambler @ 7:55 am

Steven J Vaughan-Nichol’s rant on Microsoft’s $3 software plan for developing countries is not really unexpected, given that he is a pro-Linux supporters. He covers the OLPC angle. He also mentioned the main reason for developing countries rejecting Redmond’s overture: supporting home-grown software industry better. However, what really fascinate me is this sentence in the second last paragraph

 “Dumping product is a no-no in any country’s trade plans.”

Almost every time I hear about dumping products, it is about someone violating product dumping law. The law in essence says that foreign goods cannot be sold below production price in order to drive locals out of the market. The price differential, at least to the layman, is huge. The interesting part is whether local pro-Linux campaigner can use this to blunt or stop Microsoft’s $3 initiative. From some angle, we can say this is playing dirty, after all, in most countries, Linux are imported. From other angles, it will be interesting. For example, seeing Microsoft justifying the different prices at both ed. How can one product cost so much to produce in country A but when shipped to country B cost almost nothing there? If it argues that the cost of producing an extra copy of software is virtually zero, it runs the risk of explaining to the developed world why developing world customers gets a free ride on  product development. On the other hand, if it try to says that the cost should include product development cost, how can it arrived at $3 per copy for developing countries.

Of course, we will never see this happens. Microsoft is smart. They will say the software are stripped down version. Therefore, we are comparing apples with monkey nuts. But I will still like to see an anti-dumping investigation.

A lot of commentator, Vaughan-Nichols and Larry Dignan included, pick up on OLPC. They pick up on OLPC running Fedora Core. Both did not find it worthwhile linking to earlier internet rumours about OLPC running windows, especially the rumour that one reason the SD card slot exists is for this purpose. May be I am wrong, but may be this is one of the few ways to get Microsoft Starter Edition into OLPC, if it is possible. If not, probably downgrading it for Windows Embedded, the preferred version of Windows for OLPC project as the source code is available for tinkering. Not as good as open source, but its sufficient for the aim of the project.

April 19, 2007

Strip down Windows, Office and some other MS product for $3?

Filed under: Uncategorized — ctrambler @ 2:01 pm

Before you say I am being ridiculous by offering you this idea, you should check out this blog post by Larry Dignan to convince yourself that if you are the “right” type of customer, Microsoft is indeed willing to sell you a software license for these product for a overall cost of $3. To be the right customer you must be in developing countries, and your government must agree to subsidise $1.50  towards the license.

This is of course Microsoft attempt to resegment the market to carve out another segment for developing countries in response to competition pressure. It works to reduce rampant piracy in developing countries. No longer can government relies on high cost of Microsoft product as the “justification” for not policing piracy. Microsoft is indeed extending them an olive branch by providing extremely low cost solution, abeit Windows Starter Edition is crippleware, and Office might be one as well. If I were a developing country’s government official. It’s tempting. The only question for me is is the overall payment I have to fork out, e.g. $150,0000 for 100,000 still too high a price compared to the free alternatives which allows me to get use that money to get hardware instead? The other moral dilemma that I will have is why should I subsidise something to make it affordable for my people but knowing that in the long run they probably will pay more then if the alternatives are there and cost less. Obviously, if I am corrupted, crossing my palm with silver will help!

The actual cost to the purchaser is probably going to be around $1.70, including 20 cents for backup disc which is a necessity. Even if the purchaser can burn a backup themselves, most will not do so until the unthinkable happens. (Cue Sony: Why did you just throw in back up DVD for your high end notebooks?)

I wonder what will the developed countries counterpart feels when they find out their peers in developing countries have access to cheaper products? Will it usher in the era where software companies cannot charge a premium on their software? Lets wait and see.

April 17, 2007

National Bodies Comments Are Not That Bad

Filed under: Uncategorized — ctrambler @ 8:21 am

First disclaimer, I do not know how ISO works, so if Bob Sutor says there is a lot of work to do, I tend to believe him. If Brian Jones says no as well, then I take the actual amount of work is what Sutor says and times 0.65 to 0.75. That’s my bias. I must as well admit it out front.

In reply to my previous post, Mr Jones says he will blog about National Bodies comments when it become available, so lets wait and see what he is going to say.

So, the slow wheel of ISO finally released the National Bodies comments, as they say the will do. My opinion is that the negative comments are not that bad. There are several big hills to climb, but the majority are surmountable.

One of the big hill is to justify the need for OOXML. Here, unfortunately for OOXML, as the second document format proposal to ISO, it needs  to justify itself against ODF. It is clear that a lot of National Bodies do not believe they are serving different purposes, and think that there are significant overlap. This depends heavily on the pursuasive power ECMA and OOXML proposer can muster. I will not put it past them to make their case.

Second, is the claim raised by several countries about the Patent issue. I do not really understand the issue here and the comments are too tesnse and short for me to make up my mind.

The third is perhaps a mountain. It is about missing specification on Windows Meta File. Singapore note that it is a proprietary format and claims it is hardwired into Windows OS. This point is important because it form part of the Patent problem discuss above and being proprietary and hardwired into Windows, can probably derail the whole process. However, it might be reworded as independent extension to the standard. Difficult and fraud with problems, but can be done. At the other extreme, the solution is to document Windows Meta File, but this will be difficult and definitely needs more that 5 months to do it properly.
A few countries pick up on the fact that there are undefined tags such as “footnoteLayoutLikeWW8″. In its reply, ECMA says complaining about this is like complaining that ODF have “office:config” elements. Unfortunately, this gloss over the fact that “office:config” elements are optional in ODF, exists in a different file and is normally used for different purpose. There is some arguments to be made here but it is not a big hill.

A lot of complain about the size of document and that a month period to review is too short. Norway is amused that despite the large size, Annex D is missing! We are now in the 5 month ballot process, so the size is less of a problem. Obviously ISO did not think so when it approve it for this stage.
Problems like inconsistency in using percentage are trivial to fix. Sometimes, it is trivial things like this that makes me think big gun members like British Library/Barclays Capital are sleeping on the job. I can understand that they came into the committee for different purpose than scrutinizing the actual XML syntax, but missing these are like failing to recite ABC.

Another easy to fix is Country Code, “Year 1900 is leap year” and color code. For the first and last, the simplest compromise is to adopt ISO  code alongside existing code. The same approach can be done for the second problem, using something like @date2() to represent “Year 1900 is leap year”  and @date() for “Year 1900 is not leap year” notation. It must be said, however, the more appropriate solution is for OOXML to adopt existing ISO standard. As for Microsoft, it’s simply means a bit of conversion of OOXML to whatever in memory representation of the document they do, rather than a one-to-one as it is currently. I cannot speak for Microsoft, but at least the percentage problem discuss previously, and the first and third are trivial changes, worth making to smooth the passage of OOXML through ISO.

Arguments like CGM should be used instead of Windows Meta File are and there are equivalents to OODrawingML such as SVG, in my opinion, moot. A standard is what its designer choose it to be. You do not like it and feel that it is important, then the only thing you can do is to vote against it.

Overall, surprisingly, the best prepared comment is from Kenya who seems to take the time to express virtually all the comments expressed by others, and compile it in such a way that is easily understood. Pity the failure somewhere in the PDF generation process where we did not have the figures.

Two interest comments are from Malaysia and UK. For Malaysia, it would not be interesting if not for the its national body (SIRIM) decision to sanction its working commitee on this subject. While it is not unlike IEEE sanctioning its “Son of WiFi” committee, it is regrettable that the director of SIRIM refusal to comment further rather than saying that there are serious infighting. In both cases, IEEE and SIRIM has to demonstrate that as the manager of those processes, they must demonstrate that they are “whiter than white” and impartial. Both IEEE and SIRIM claims that the committees are too biased. In the case of IEEE, IEEE went into specifics by saying that rival proposals are not rejected out of hand. In the case of SIRIM, the director did not give example of how ODF opponent’s view were suppressed. He simply give a broad statement that they are. Both IEEE and SIRIM sanction the committee but do it differently. IEEE replace the chairperson but leave the committee more-or-less intact, which is better than SIRIM’s decision to replace the whole committee. By removing the chairperson only, IEEE send a strong signal that everyone involved must work collaboratively as required by the charter of the committee and did not seek to unduly influence the composition of the committee to influence the outcome, but SIRIM’s decision to replace the whole committee open itself up to charges of unduly influencing the composition of the committee to influence the outcome and looks heavy handed. No doubt the new SIRIM committee will bear some resemblance to the old, but why not just insist of replacement of the troublemakers?  Before we read too much into it, SIRIM is unhappy about ODF adoption process in Malaysia. It is related to, but not the same as its comment on OOXML. I was hoping that SIRIM’s comment here will shade some light.Unfortunately it did not. Its comment is one of the better argued comments of the lot.

UK is interesting because Microsoft started an online petition. Interesting because it is the only country such a petition exists. It could mean Microsoft feels that it needs to get UK on its side and is not getting it. UK’s entry stand out because like classic good work from UK government and statutory body, they based all their comments by citing all the authorities they rely on to make the comments and where appropriate, list out what their standard are when judging the proposal. It also show their mastery of the language by providing the best organized document. Furthermore, they call a cat a cat and do not mince their words. For example, they says what others did not or dare not say, that it is their opinion that “N8455 [OOXML] cannot be fully understood or implemented by a typical computer programmer  without substantial technical assistance from Microsoft”

Next Page »

The Rubric Theme Blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.