LWN: Comments on "Obama Inauguration shines on Linux too with Moonlight (ars technica)"

You need more examples?

man_ls — Tue, 27 Jan 2009 23:11:36 +0000

The "portable" NT operating system (x86, Alpha, a host of other architectures coming soon) comes to mind. Or the Mac OS port of Office -- discontinued at the loss of substantial revenue, though later revived by a Microsoft held hostage by Apple patents.

You want to keep thinking Microsoft will not discourage (discontinue?) the Linux port of an emerging technology once it is not "strategic"? Be our guest.

Parallels? You want parallels?

khim — Tue, 27 Jan 2009 22:58:27 +0000

Even in this dominant position Flash has been sort of forced to pay attention to Linux. After all we do have a binary only Flash player from Adobe... this means that Linux cannot be ignored even by the strongest player in that field.

And Netware had support for DOS, Mac, OS/2 and few versions of UNIX (eventually including Linux via Caldera).

So, Silverlight wants to displace Flash... is it possible doing that, then killing Moonlight, and still staying ahead of Flash?
Maybe yes, but I doubt it.

It's a gamble. The biggest problem for Microsoft is the fact that Flash is supported relatively well on many platforms. There are even planned port of "full-size Flash" for mobile platforms! This is the biggest problem for Microsoft. But Silverlight can not solve this problem while Flash is the king. So they are playing "we are open" card - just like they did with COM/DCOM (remeber press-releases like this one or this one?), OpenGL (reme ber how it was?) and other "open" technologies... When market will be captured - phase two begins. This is tried and true approach and it was documented many times and even explained in details in mails from Microsoft itself (revealed by court)!

I really hope that now that we have the specs Gnash will leap forward, but the truth is, Moonlight has the specs and a sponsor, so it's still in a better position

Of course! Microsoft needs to kill Flash fast - or it'll have no real features to introduce in phase two, when Windows will go forward and all other systems will be left in dust! The fact is: for Adobe Flash Player is not a central product. And the fact that it's crossplatform is pure plus - the only question is "does money spent developing Linux version are worth it or not". Gnash does not endanger it's cash cows. For Microsoft Moonlight is real threat directed to the heart of the empire. It makes Windows less relevant. Thus they MUST kill it at some point - or else the whole empire is in danger. To think that they'll support this gamble till the bitter end is the height of folly! The only real question is: how long will they continue with this charade. If Silverlight will be unable to capture market fast enough then Microsoft will just abandon the whole technology and introduce something new, but you can be pretty sure good crossplatform support is not something they want to have in the end...

I've chosen wrong victim of Microsoft

massimiliano — Tue, 27 Jan 2009 18:07:56 +0000

OS/2 situation was indeed different: Moonlingh's game look like CIFS's game. Microsoft uses many plays to leverage it's monopoly and kill the competition, plays are slightly different...

Oh, well, as I said I have no crystal ball.
But there are still elements that make me think I'm realistic in wishing differently...

For start, Silverlight is competing with Flash. Now, for several years Flash had no competition at all: it squashed Java applets easily, and now there's Silverlight, but in the middle it has been the king of the hill (and it still is).
Even in this dominant position Flash has been sort of forced to pay attention to Linux. After all we do have a binary only Flash player from Adobe... this means that Linux cannot be ignored even by the strongest player in that field. So, Silverlight wants to displace Flash... is it possible doing that, then killing Moonlight, and still staying ahead of Flash?
Maybe yes, but I doubt it.

Then, another observation: since Silverlight sort of "needs" a Linux presence... IMHO, it's much better having Moonlight (with soource code, and a team paid to develop it as Free Software) than a binary only plugin. I really hope that now that we have the specs Gnash will leap forward, but the truth is, Moonlight has the specs and a sponsor, so it's still in a better position. And this, regardless of what the future will bring, is still a good thing now.

I've chosen wrong victim of Microsoft

khim — Tue, 27 Jan 2009 14:49:41 +0000

OS/2 situation was indeed different: Moonlingh's game look like CIFS's game. Microsoft uses many plays to leverage it's monopoly and kill the competition, plays are slightly different...

Ok, may be I was wrong

khim — Tue, 27 Jan 2009 14:47:06 +0000

Your comment is insightful, however IMHO the analogy is flawed, like comparing apples and oranges.

You are right: situation with Mono is not quite like situation with OS/2 and Moonlight's situation is totally different. It's more like Netware situation: Netware was "king of LAN" in 80th. And Microsoft knew it's be hard to compete with the leader alone so it proposed new standard (CIFS) in 1996, even submitted it to IETF (thankfully it was rejected), etc. Few years later... standards are no longer needed as Windows NT is firmly established as the "LAN server". It's time to kill other competitors! And now we have this mess with Active Directory and such.

Thus tes question regarding moonlight is not "will Novell be screwed agaian?" but "when will Novell be screwed?". And looks like the answer is: pretty soon. Looks like Microsoft planned to do it this year but somehow Silverlight was not as successfull in killing Flash as Windows NT was in killing Netware server thus we have article which praises Microsoft for postponing this massacre. It's sad, really: looks like people never learn...

Up to now Moonlight is likely helping Microsoft in raising Silverlight adoption: Linux has a desktop share similar to that of the Mac, so Microsoft cannot go against Moonlight: it would harm Silverlight. And if (or when!) we will get to the point that Linux has gained a desktop share so large that Microsoft gets scared, and thinks of eliminating Moonlight to try to harm Linux, this will again make no sense: what harm will it do to Silverlight?

Yes, that's one possibility. If Linux will gain popularity faster than Silverlight then it'll be the outcome. If Linux will not gain popularity as fast (and it looks like it will not) then Microsoft will have time to kill Flash first and then return back to Linux.

In other words: nobody has a crystal ball, and nobody knows what exactly will happen in the next years. But the situation is definitely not the same as the one of OS/2 vs. Windows in 1994.

You are right, of course. It's closer to Netware Server vs Windows NT Server in 1995... Not a big difference IMO...

That's how OS/2 was killed...

man_ls — Tue, 27 Jan 2009 07:00:46 +0000

Of course a lot of details differ, but I think the analogy is still significant. There is one party helping a group of developers to ensure compatibility, as long as it is in that party's interest. Afterwards the help is retracted and the developers cannot stay compatible for long.

And of course the party is the same: Microsoft. Maybe their interests stay constant, but a slight change in their priorities and feature parity (even now a major release behind) would soon be lost.

The advantage of Flash in this respect is that it is a mature platform and has a huge installed base. Once something starts working it is not very likely to change. And now with a public specification the situation should be even better, since implementors do not rely on Adobe. The spec could also be retracted, but that is not how Adobe has behaved in the past (e.g. with PDF).

That's how OS/2 was killed...

massimiliano — Mon, 26 Jan 2009 16:59:12 +0000

Your comment is insightful, however IMHO the analogy is flawed, like comparing apples and oranges.
I try to explain...

The first mistake is comparing OS/2 with both Mono and Moonlight. Mono and Moonlight, while sharing pieces of technologies (the .NET runtime in the case of Silverlight 2), are very different "products" with very different targets, so the idea of "success" and "failure" is different for each of them.

Let's compare OS/2 and Mono first: the analogy makes sense as long as you look at them as two development platforms. A development platform is healthy as long as there are people writing software for it, so that you have applications available for the platform.
The fatal problem with OS/2 was that it never gained share as a development platform of its own: people just considered it "a better Windows than Windows 3.1", and when Windows went ahead it was left behind (I have been working in IBM between 1994 and 1995 so I remember it clearly).

The situation with Mono is completely different: Mono was born as "a better development platform for Free Software", in the sense of "better than C and GObjects", "better than Java", "better than Python and GTK bindings"... we can debate if it's truly technically better, but it is gaining a user base (among developers) that has nothing to do with Windows. It is used for Linux desktop software, it is used for server software, it is used in embedded systems... it is also ending up being a wonderful platform for programming the logic of videogames, used also on the iPhone and the Wii console. This has nothing to do with Windows.
Mono is benefiting immensely from the fact that it uses the same standard as .NET because there are a lot of developers that know it, and because there are a lot of good books that explain it, but Mono is definitely a different thing, and does not need Windows to walk on its own legs.

And yes, Mono also offers a "Windows compatibility story", but nobody has ever said that it is complete or that it will ever be. In certain areas it is complete, and it will be kept up to date as long as it seems useful, and there are people benefiting from this.
But it's not the main purpose of Mono, and has never been: if it becomes a problem Mono can live without it (something that OS/2 did not manage to do).

Then, let's compare OS/2 and Moonlight: Moonlight's goal is mainly to be a Silverlight replacement for Linux/Unix, but in this sense it is not "competing" with Silverlight (while OS/2 was competing with Windows). Actually, it is maybe helping Silverlight making it really cross platform, so the scenario "Silverlight kills Moonlight" does not really make sense, certainly not in the same sense that "Windows killed OS/2".

If you think about it, Moonlight is the official Silverlight implementation for Linux, and it is Linux that competes with Windows, not Moonlight that competes with Silverlight. In the case of OS/2 its windows compatibility was so important that killing it was enough to kill the whole OS, but this scenario does not apply to Linux.

Up to now Moonlight is likely helping Microsoft in raising Silverlight adoption: Linux has a desktop share similar to that of the Mac, so Microsoft cannot go against Moonlight: it would harm Silverlight. And if (or when!) we will get to the point that Linux has gained a desktop share so large that Microsoft gets scared, and thinks of eliminating Moonlight to try to harm Linux, this will again make no sense: what harm will it do to Silverlight?

In other words: nobody has a crystal ball, and nobody knows what exactly will happen in the next years. But the situation is definitely not the same as the one of OS/2 vs. Windows in 1994.

On the other hand, whether we like it or not Silverlight is a "web technology" that, being backed by Microsoft, is here to stay, and will be used on the web. With Moonlight we have a Free implementation which makes it "more free than Flash", and that IMHO is a good thing for desktop Linux.

Again, IMHO, this is much better than having no Moonlight at all and ending up in a situation worse that what we have with Flash...

Are you sure? How?

khim — Mon, 26 Jan 2009 15:45:02 +0000

Seriously, look outside the ODF kerfuffle and investigate standards in other areas of engineering.

Sure. How about this example?

Yes, it's true that sometimes more then one standard is approved for one field - but this route should be chosen only when alternatives are worse...

That's how OS/2 was killed...

khim — Mon, 26 Jan 2009 14:18:09 +0000

Do you remember OS/2? This platform had perfect emulation of Windows 3.1, it passed "no regression test suite" test suite, IBM claimed it was 100% compatible with Windows (and it was almost true) and... that's what killed OS/2. Nobody had any incentive to test it's creation with OS/2 (it had perfect Windows compatibility) so companies started to create and test stuff only with Windows - and when Microsoft "pulled the plug" (Windows 95 was not included in Microsoft/IBM agreement) they just dropped OS/2 support without any though.

The same can happen with Moonlight and Gnash - but where with Gnash people expect incompatibility and are ready to fight it (if they do care about Linux at all) with Moonlight and Mono situation is very different: they were promised "perfect compatibility" and so when Mono 2.2 is released they compain that it's unusable because there are no WPF instead of being happy with AOT.

Basically if you go with "full support" model - you need simultaneous releases of different versions (and the fact that Novell developers are doing the job, and they do not report to Microsoft managers guarantees that this approach will fail), if you go with "we are independed, not 100%- compatible implementation" - people are forced to care.

Sorry but when someone violates law it's not a proof that law does not exist

NigelK — Mon, 26 Jan 2009 09:53:59 +0000

That would be about making sure that standards are implemented properly, not that standards do not overlap one another.

Seriously, look outside the ODF kerfuffle and investigate standards in other areas of engineering.

Microsoft can pull the plug at any time...

massimiliano — Sun, 25 Jan 2009 20:32:07 +0000

Well, if a stable release of Moonlight does not work on a site where Silverlight works, it's a bug that will be handled, you can be sure of that.
Just take care of reporting the bug :-)

Moonlight is anyway in a different position wrt Gnash: Moonlight is required to pass the same no regression test suite that Silverlight passes... believe me, it makes a difference.

Gnash is not controlled by Adobe. Moonlight is controlled by Microsoft (via Novell).

The only "control" that Microsoft has on Moonlight is that Microsoft requires (in the cooperation agreement) that Moonlight works effectively as a reimplementation of Silverlight. Being able to pass the test suite is related to this requirement.

Apart from that, Microsoft has no control in how Moonlight is done. Novell developers are doing the job, and they do not report to Microsoft managers.

But the most important thing is that Moonlight is Free Software, and in this sense it is not controlled by anybody: those that do it dictate how it's done, and everybody is free to fork it and go on from there in a different direction. I guess that Gnash works in the same way.

I really don't see how Microsoft controls Moonlight: the requirement that Moonlight is a 100% compatible replacement for Silverlight does not seem a bad thing to me given the goal of the project.
And if you don't like it, go ahead, take the code and fork it: Microsoft will not control you any more than how Adobe controls Gnash :-)

Sorry but when someone violates law it's not a proof that law does not exist

khim — Sun, 25 Jan 2009 09:43:48 +0000

Heck, not even the various standards bodies insist on there only being one way to do things.

Actually they do. ISO's motto was "One standard, one test accepted everywhere". Of course money talk and now we have stupidity like ODF/OOXML, but it does not mean the whole principle is bad. Flash was bad (when we had SVG) but Silverlight is ten times as bad: Flash replaced non-working (even if W3C-blessed) solution, Silverlight is trying to replace working solution.

Fighting inercia

man_ls — Sat, 24 Jan 2009 12:05:41 +0000

I will add a few notes to khim's apt reasonings.

It is sad, but the areas of human expertise are by necessity very small; and inercia is both a necessary evil and a requirement for certain activities. This last point is what you seem to fail to grasp. I may bravely perform a lone rethink of our entire History and find out that agriculture was a big mistake; and go with my herd of sheep to graze in your parking lot. Guess what: owners will not appreciate it and probably my enterprise will rapidly fail. (Believe it or not, something like this happened in Spain in the 17th century to some government officials, according to Brennan: they wanted to reallocate a big population of famished farmers to sheep herders. And in the process ruined a big part of the Castilian economy.)

The consequences of human acts are not easy to predict. Sometimes well-meaning changes bring a lot of pain to the world. Solutions that involve changing everything are often more perilous than the mere inercia they fight. Revolutions do happen, but forcing them has not yielded good results. That is why (we) armchair strategists, social reformers and legislators pose such a danger to Mankind. Sometimes there is no alternative but to let specialists do their work, and see what happens. And then judge the results.

In the case of Microsoft we can think that they are evil and should be disbanded by the government. That is what the US government tried and failed. We can also blame people for buying their shoddy products. Guess what: Windows still outsells the competition by a wide margin, even in side-to-side competition as with laptops and netbooks. But somehow the attitude of the UE, bland as it seemed at the beginning, is bearing fruit: they imposed huge fines for opaque protocols and dirty tactics. Microsoft is now cooperating with outside entities and helping them to interoperate. You have to at least appreciate the results.

Groupthink? Give me this kind of groupthink any time of the day.

What's the difference?

khim — Sat, 24 Jan 2009 01:02:43 +0000

But much of my posts refute that certain behaviors are groupthink, despite man_ls claims that my statements lead to those conclusions (twice).

I don't see any credible refutes.

There are phenomenon in human society: small group dictates it's will to the rest of society. It's agriculture (small number of agriculturists develop farming rules, farmers follow them), medicine (small number of pharmaceutists develop new medicine, doctors follow rules), modern physics (very few scientists develop stuff, chineese workers make PDAs, laptops and so on), Linux kernel (small group of kernel developers make a kernel, millions use it and never dissect it), Microsoft monopoly (Microsoft uses it's money and power to push it's technology, people are blindly accepting it).

Why the first examples are not "groupthink" while the last one is? In other words: why famine from poor choice of crop is agriculturist fault while Microsoft's abuse of it's power is somehow fault of those who push and use the monopolists' products and anything that requires them?

I'm yet to see any explanations - just repeated "huh?" questions. If you place blame for wars on "agreeable soldiers" then why you are not putting the same blame for destroyed ecology in China on yourself? Certainly if you'll stop buying stuff made by poor chineese workers ecology will be much better there!

However, please do not attribute those conclusions to me, especially since I argued against them.

I never attributed these conclusions to you - but your original statements certainly lead to these conclusions...

Suppose there are simple yet tragic story: serial rapist is using the same scheme again and again - talk with a girl, buy her a present, bring her to his apartment, then rape her. This pattern was repeated many, many, many times. And you are seeing as this same rapist is being mushy-mushy with yet another attractive girl. What will be your first reaction
1. Oh, how wonderful - we'll see new happy marriage soon... or...
2. Uh-hu - the next victim is here, I see...

Microsoft is certainly performed that trick enough times (CIFS, Fahrenheit, COM/ DCOM, etc) that I don't think about possibility #1 anymore - and this has nothing to do with groupthink. Microsoft is only nice to the competitors when it's underdog or is under intense pressure (from EU or others). May be someday it'll change - but not while Ballmer is CEO...

Sorry?

martinfick — Fri, 23 Jan 2009 23:49:17 +0000

I don't see where I said anything like what you claim I said. I know this is a long thread, perhaps you got someone else's replies mixed in with mine? I followed your second link to my supposed statements and it doesn't even seem to be the same tangent of this thread than the one you are talking about. I could be wrong, perhaps you left out some explanations you thought were clear?

But to be clear what I said/believe: I do believe there is much groupthink in many fields. I certainly do not think that it is OK. But much of my posts refute that certain behaviors are groupthink, despite man_ls claims that my statements lead to those conclusions (twice). Now, if you disagree with me and believe that my original statements lead to the same conclusions as man_ls did, there is probably not much I can do to change your mind. However, please do not attribute those conclusions to me, especially since I argued against them.

I also do not believe that you can make full investigations about everything, but you should reason about everything. And, if you see evidence that disagrees with "common knowledge", perhaps you might question whether this "common knowledge" is accurate? If it is not, perhaps groupthink or some other flawed logic (or lack of logic) led to this false "common knowledge". But the truth is, this thread (when I entered it) was not about "knowledge", but rather very subjective opinions about MS's motivation for certain behaviors, something which is much more prone to groupthink. Extrapolating (incorrectly) my statements to scientific progress and buying a Honda Civic can be fun, but please pay attention to which parts of the conversation where actually mine before attributing opinions and statements to me.

"Sorry, but no, it's megacrime:"

What are you claiming is megacrime?

Microsoft can pull the plug at any time...

khim — Fri, 23 Jan 2009 23:10:32 +0000

Which more or less work, more often not...

That's the expirience with Silverlight, too...

Or you are implying that Moonlight is in some way less Free than Gnash?

Yes. Gnash is not controlled by Adobe. Moonlight is controlled by Microsoft (via Novell). Granted, it's not very short leash, but it's a leash...

Sorry?

khim — Fri, 23 Jan 2009 23:04:32 +0000

True, most of the public has not personally done the scrutiny, because they trust the experts -- and that is good.

Sorry, but no, it's megacrime:

If you merely switch your group think from the MS loving group to the MS hating group you will still be using group think.

If you'll try to even learn all things required to actually built LHC - you'll stumble upon small problem: human life is not long enough. So there are always groupthink in agriculture, medicine, modern physics or the Linux kernel... Why it's bad here are not there?

You don't personally scrutinise every engineering detail of your car, either, but are willing to believe based on others' reviews and reports that the Honda Civic (say) is a reliable car. No human mind can encompass all the expertise required to personally verify every claim encountered in life.

You are correct, of course, but from what martinfick is saying it's Ok to do so in some cases (agriculture, medicine, modern physics or the Linux kernel) but not in other cases (Microsoft hating). Why? What's the difference between Microsoft lover/hater and car buyer?

Apologist at work

martinfick — Fri, 23 Jan 2009 19:37:54 +0000

I guess if you really want to blame both, go ahead. :) But, it's a fine line between a salesman and a con artist. In general it makes more sense to hold those making bad personal decisions responsible for them then the person who talked them into it. Depending on the level of con, of course.

Naive people become very unnaive pretty quickly once put through the ringer. Overprotected people continue to be overprotected when over protected. If you blame the con artist, where do you stop? Do I blame my teachers for not teaching me how to avoid the con artist? Do I blame my parents for bringing me into a world full of con artists? Or, as most, do I simply blame the government for not stopping the con artists? Personal responsibility empowered by personal reasoning is the sane solution.

Just because it is XML?

massimiliano — Fri, 23 Jan 2009 19:27:02 +0000

At least for Flash there are a couple of open viewers not under direct control of Adobe.

Which more or less work, more often not...

With Silverlight we have Moonlight, which, while developed by Novell, is Free Software just like Gnash and Swfdec. Or you are implying that Moonlight is in some way less Free than Gnash?

Obama Inauguration shines on Linux too with Moonlight (ars technica)

massimiliano — Fri, 23 Jan 2009 19:21:18 +0000

AFAIK, you must download the official blessed binary blobs if you actually want to use it in the real world.

This is false: you can recompile it from source and use ffmpeg as decoder, and it will work just fine. The binary codecs are there only because the status of ffmpef with respect to patents is problematic, but if using ffmpeg is ok for you, just use it!

Apologists against apology

rsidd — Fri, 23 Jan 2009 09:51:56 +0000

I understand "groupthink" to mean "mass acceptance without question". Scientific ideas, the Linux kernel, etc have been extensively scrutinised before acceptance. True, most of the public has not personally done the scrutiny, because they trust the experts -- and that is good. You don't personally scrutinise every engineering detail of your car, either, but are willing to believe based on others' reviews and reports that the Honda Civic (say) is a reliable car. No human mind can encompass all the expertise required to personally verify every claim encountered in life.

On revealing oneself...

xoddam — Fri, 23 Jan 2009 07:32:54 +0000

"I am a woman." -- Andrea Dworkin, on being asked if she was a lesbian.

Obama Inauguration shines on Linux too with Moonlight (ars technica)

salimma — Fri, 23 Jan 2009 03:47:48 +0000

Only very recently; the complete spec (which takes some hunting to find, last time I looked) has not been out long enough for Gnash and swfdec to be completely usable. Whereas Moonlight is pretty much fully Silverlight 1.0 compatible, and will be Silverlight 2.0 compatible in a few months' time.

Apologist at work

lysse — Fri, 23 Jan 2009 01:46:41 +0000

Isn't that rather like saying "don't blame the conman, blame all the gullible people that allowed him to rip them off"?

too late.

ncm — Thu, 22 Jan 2009 23:25:02 +0000

... openSUSE (my preferred platform)

You reveal yourself.

Apologist at work

Wol — Thu, 22 Jan 2009 23:14:33 +0000

"Illegally buttressing its monopoly"

MS puts loads of code in to make sure minor programs work. There's also a fair bit of evidence they put code in to make sure their major competitors DON'T work. Like WordPerfect, for example.

Why, given MS's alleged approach to making sure that programs run, was installing networking onto Windows 3.1 a *sure* *fire* way of making sure WordPerfect suddenly became extremely crash-prone? Why did installing Office 95 over WP6.1 result in WP never working again? (The ONLY fix I found was to wipe Windows 95 and reinstall from scratch!)

Cheers,
Wol

Just because it is XML?

man_ls — Thu, 22 Jan 2009 21:47:58 +0000

Just curious, why do you think those technologies are better? XAML is XML, yes, but to me it is as opaque as an .swf file. At least for Flash there are a couple of open viewers not under direct control of Adobe.

Obama Inauguration shines on Linux too with Moonlight (ars technica)

cortana — Thu, 22 Jan 2009 15:38:25 +0000

Is it really more open? AFAIK, you must download the official blessed binary blobs if you actually want to use it in the real world.

Apologist at work

pboddie — Thu, 22 Jan 2009 14:56:32 +0000

No matter what your opinion of governments, its simply illogical to ever think that they could help fight monopolies (without replacing them with bigger/stronger monopolies).

Here's how governments can help: instead of insisting that taxpayers and members of the electorate purchase proprietary software from a single company in order to interact with agencies of that government, under threat of criminal penalties should one fail to interact with such agencies (by, for example, not managing to declare one's taxable income), governments should at the very least employ open standards in order to give their citizens the choice of which software they may use, and I would also argue that Free Software implementations should exist and be provided free of charge for interacting with government agencies using such open standards. If someone wants to pay for the thrills of using Microsoft TaxFiler Deluxe, they should be free to do so, but those who see no need for such "luxuries" should be able to carry out their obligations without additional burdens.

As a consequence of pursuing a policy of openness and interoperability, governments would not, either through laziness or through corruption, be contributing to a situation where a single company's products are effectively mandated and where the costs of such products are imposed like an additional tax on the electorate. Every time some agency insists that a Microsoft technology be used to take advantage of its services, a monopoly is being strengthened and the mutually reinforcing excuses of "you'll be running Windows anyway" and "nobody wants to run anything else" are perpetuated. Yes, governments can not only fight monopolies - they can create and reinforce them, too.

Apologist at work

pboddie — Thu, 22 Jan 2009 13:09:03 +0000

Yes, of course, never doubt that they're *EVIL*, just *EVIL*!!! Despite all evidence (like the above and their excellent assistance to the Samba team)... *EVIL*.

Leaving aside any observations that some people are easily satisfied and stop asking pertinent questions when a binary RPM for their architecture shows up for the shiny proprietary technology of the day, it should be noted that "their excellent assistance to the Samba team" would never have happened had it not been for persistent pressure from the European Union. Portraying the situation as you have done is a bit like noting that "Johnny plays so well with all the other children! How can you say he's a bad boy?" while omitting to mention that the headmaster has threatened to expel him from school for a variety of anti-social behaviour.

Apologist at work

pboddie — Thu, 22 Jan 2009 12:56:20 +0000

Maybe it's time to stop blaming the "monopolists" and to place the blame squarely on those who push and use the monopolists' products and anything that requires them?

Well, I don't disagree that those pushing Microsoft products often do so out of convenience. It's difficult to go to, say, an electrical retailer and tell them that they should offer products which don't run Microsoft Windows, especially when they can just say that there's "no demand" for anything else and that even though some people have asked for alternatives, there's no way to really test that demand without offering such alternatives, and that would involve effort which might not pay off.

Really, maybe it's time to stop blaming leaders for wars and to realize that it takes agreeable soldiers and institutions to fight them? If you merely switch your group think from the MS loving group to the MS hating group you will still be using group think.

Who said that I was in any group? I don't use Microsoft products and I deliberately purchased hardware that didn't incur the "Microsoft tax", which I think you'll find, given the lengths people need to go to in order to make consumer laws and regulations apply to Microsoft and vendors so that they may claim a refund for an unwanted bundled product, is a matter which does need regulatory attention.

And it's not that I don't use Microsoft products merely because they're made by Microsoft as you patronisingly suggest. It's because I don't agree with the licensing of those products, the associated regime of control that the licensing imposes, or the anti-competitive behaviour of the company which produces them; I have no time for Apple products, either. I think you'll find this to be a satisfactory example of me exercising my "personal reasoning".

too late.

dgm — Thu, 22 Jan 2009 10:23:12 +0000

> Yes, I'm one of those annoying pragmatists who uses effective technology. Fortunately it works well on openSUSE (my preferred platform).

A true pragmatist is well aware of the pitfalls when a compromise has to be made. Lazy people on the other hand, happily follow the easy path.

Obama Inauguration shines on Linux too with Moonlight (ars technica)

NigelK — Thu, 22 Jan 2009 09:45:35 +0000

"Why do we need this? Flash works perfectly fine for me."

There's no reason why there can't be more than one way to do things. Heck, not even the various standards bodies insist on there only being one way to do things.

Feel free not to use it, but refrain from preventing others from taking advantage of it.

Apologists against apology

man_ls — Thu, 22 Jan 2009 08:09:27 +0000

Sorry, I was too terse. My point is that for every field listed, even if key inventions were the result of individual work, true progress could only be achieved when common people sheepishly accepted the ideas that a few pioneers had put through.

The case of agriculture is prototypical. To achieve abundant crops the sumerians (and later the babylonians) had complex irrigation systems that very much guided all of their civilization. A lot of concepts had to be invented, such as that land is "property" and only the owner can profit from it; it can be owned and sold. There were big tensions with the surrounding cultures, basically nomads which came and went causing damages in the process. Echoes of these tensions are there in westerns, where farmers and herders fight each other: herders feel that land should be free for them to roam, while farmers defend their properties. In the end farmers always win -- because of what you call "groupthink". Nowadays the concepts are so ingrained into our "groupthink" that my reasoning required some explanation, but they are there. Trespassers are not allowed in my property, and your cattle is not welcome to graze in my parking lot.

For science it is very much the same. After the advances of previous centuries, scientific progress of the kind achieved through the 20th century can only be achieved with government funding. It is a collective achievement in the sense not only of thousands of individuals cooperating freely, but many hundreds of millions funding them. Maybe for individualists it is hard to accept, but the synchroton, atomic energy and deep-space cosmology are not the result of a few heroic individuals -- they were encouraged and funded by large government agencies, and a public belief that they were doing good.

Let's not even get into large companies since it would take too much effort. But likewise for the Linux kernel. People don't get into it questioning the basic ideas -- monolithic vs microkernel is only the most visible and still debated issue, but things like programming language, changing module APIs, license or code inclusion are mostly accepted as is by the principal maintainers. If every contributor were to discuss every little issue no progress would be made at all (and even now it is pretty hard by what they tell), or everyone would start their own kernel project. No, "groupthink" makes kernel devs join and prod along -- and the rest of us to encourage them and look at them with respect.

If "groupthink" is made to mean, as you imply, just "received notions accepted without question", then it is a mixed bag: full of good things with the occasional snake. The regular meaning of "groupthink" as "notions accepted by a group that defy common sense" looks much more useful to me. In fact, the former can even be counterproductive: since questioning all received knowledge is, by definition, impossible (we would be arguing every word, since languages are also received knowledge), then individualists have to focus on a select group of received ideas. These conflictive ideas are sometimes picked at random, but at times they are chosen for them by a smaller group of individualists which then stablish, by "groupthink", what they have to fight. You gotta love paradoxes.

Apologist at work

Mithrandir — Thu, 22 Jan 2009 01:15:06 +0000

It gladdens my heart to know that people at MS love FreeBSD so much... :P

Apologists against apology

martinfick — Thu, 22 Jan 2009 00:01:41 +0000

If we follow your reasoning, agriculture, medicine, modern physics or the Linux kernel are nothing but the result of "Group Think".

I don't see anything in my reasoning that says those things are group think. Please explain? Quite to the contrary, most of those things were the result of individuals who reasoned against common accepted "knowledge", some of it even "group think"! :)

Often times individuals reasoned and suggested "preposterous" things that others (sometimes because of groupthink) denied and argued vehemently against until finally more and more people slowly accepted these preposterous ideas because of the evidence involved. This is this exact opposite of group think. I certainly was not suggesting that simply because theories actually stood the test of experimentation (not political talk) and became accepted on a widespread scale, that these theories were group think!

In fact they have allowed a bunch of very few people (much smaller than 20%!) to impose their views on the rest, and to make them rearrange accordingly.

Where is the imposition? Science itself can't impose, only government mandated propaganda/science can be imposed. You have the right to reject those things that you mention above if you want, they are not imposed on you. Or, am I missing your point, is that what you really meant?

Why do we all have to pay for the LHC or the Hubble telescope?

Is that really a question, sounds more like an implication that scientific progress is only made by forcing us all to pay taxes for government funded scientific research?

Why is land "property"?

Huh? Please explain this question and its relation to the conversation?

Why do large companies even exist?

Again, huh? I assume that is some sort of reference to corporations being a mechanism to support group think? Yes, certainly they can do that, which would certainly explain why many people fear and hate big corps, don't you think? But, at least they are not more powerful than the sum of the individuals that support them, so the corporation mechanism is not as amplifying to group think as most democratic "force backed" processes (governments).

Or why do we have a monolithic design in our kernels? Groupthink!

Wow, very strange. I can't see how a monolithic vs. (what?), microkernel, design relates to groupthink. I am sure it might be possible, but you really will have to explain that one. To think that you got that impression from my post, well I am sorry if I was that unclear to send you so far off track from what I was suggesting.

My personal impression of linux kernel development processes (mostly gained from LWN coverage) is that groupthink is a rare thing and that personal reasoning is very prevalent throughout the process. Perhaps that is why some people prefer the "benevolent dictator model" to a more democratic one? (I am not claiming this, it's just an idea that this line of thinking brought me to.) In the linux process you almost have a hierarchy of benevolent dictators, each one requiring solid reasoning (and often even evidence) before they will accept new ideas. This probably acts as very effective multi layer groupthink filter! Most of these "dictators" will not accept code simply because some large groupthink mechanism, such as a corporation, or a slashdot thread, ;) insists on it.

Obama Inauguration shines on Linux too with Moonlight (ars technica)

Burgundavia — Thu, 22 Jan 2009 00:00:48 +0000

My comment was from the end users perspective. And the Flash spec is completely open.

Obama Inauguration shines on Linux too with Moonlight (ars technica)

whitemice — Wed, 21 Jan 2009 22:56:01 +0000

If you look at the technologies Silverlight (XAML and .NET) vs. Flash there is little comparison. Not to mention that Silverlight is dramatically more Open.

Apologist at work

dlang — Wed, 21 Jan 2009 22:50:47 +0000

the key problem is in your statement

Microsoft works well with others when they are forced to.

and then you soften it a bit by adding

When it's their self interest.

it needs to get to the point where Microsoft considers it in their self interest to work well with others period, end of statement.

as long as they keep fighting against working with others, to the point that they have to be forced to, they won't be trusted

microsoft has gone back ond forth on CIFS

initially they were happy to push it as an open standard, then when SAMBA started to threaten them they tried to make it proprietary, then they were force to open the specs, so they did so (but only if you were willing to pay them $10,000), now I think I've heard that they are easing that.

hardly the actions of someone being cooperative. these are the actions of someone looking for any advantage at the expense of everyone else. (and another symptom of the problem)

In the opensource community we do work well with competitors. what we don't do is give control over things that we depend on to our competitors.

you say that if Microsoft kills flash you expect them to try and kill moonlight, so why are you surprised that others who believe the same thing aren't willing to help microsoft kill off flash?

cooperating with competitors who will cooperate with you is one thing. handing the mugger a knife so that he can stab you with it is a completely different thing.

too late.

whitemice — Wed, 21 Jan 2009 22:49:31 +0000

Yes, I'm one of those annoying pragmatists who uses effective technology. Fortunately it works well on openSUSE (my preferred platform).