Moonlight on 04 Jan 2008
Ville Voutilainen
technologies. People are understandably wary. I would not list
IronPython as a thing that Microsoft open-sourced, quite the opposite.
The original IronPython was AFAIK available under a bsd-style license,
when the author of IronPython joined Microsoft, IronPython was
released under some sort of shared-source license. Minor step
backwards IMHO.
ara...@gmail.com
"I personally want to write cross platform web applications using C#,
Boo, Python and Ruby. And that matters to me,"
I think, regardless of all the politics and the fuss Mr Icaza is
creating with his controverse dicisions, what is very good about them
is their honesty and clearness.
rram...@gmail.com
The fact that the technology exists and might be used by everyone -
the true meaning of interoperability - is the most important fact in
all this.
Completely agree with your post.
greg
"In addition to the Silverlight release, we've also made the full
source code for both IronPython and all of the new DLR platform code
available on codeplex under the BSD-style Microsoft Permissive
License. All of that code can be downloaded today as part of the
IronPython project at codeplex.com/ironpython."
See: http://blogs.msdn.com/hugunin/archive/2007/04/30/a-dynamic-language-runtime-dlr.aspx
The MS-PL is an OSI certified license see here:
http://opensource.org/licenses/ms-pl.html
From Microsoft:
"Microsoft Public License (Ms-PL) - The Ms-PL is the least restrictive
of the Microsoft source code licenses. It allows licensees to view,
modify, and redistribute the source code for either commercial or non-
commercial purposes. Under the Ms-PL, licensees may change the source
code and share it with others. Licensees may also charge a licensing
fee for their modified work if they so wish. Microsoft uses this
license most commonly for its developer tools, applications, and
components."
See: http://www.microsoft.com/resources/sharedsource/licensingbasics/sharedsourcelicenses.mspx
Seem's like open source to me - or am I missing something?
Cheers,
Greg.
On Jan 4, 7:32 pm, Ville Voutilainen <ville.voutilai...@gmail.com>
wrote:
tim
it, extending it, innovate with it, prototype with it, and enter the
same cycle that Linux, or web browsers have entered? Or someone
turning it into a standard?
Nothing. "
Patents?
Vexorian
proprietary ways of making the web much more bloated with animations
and other silly unnecessary things.
Promoting moonlight right now does nothing useful for Linux,
Silverlight is simply not used yet, the only thing it does is helping
Microsoft advertise their attempt of taking the web over again.
Something I really cannot agree is to see the web being hijacked again
by Microsoft and... this time with an extra .net requirement. It just
does not make any sense to promote moonlight so.. early, unless of
course you are Microsoft or you think Microsoft taxing the web is a
great idea.
Regarding your last question, these are things that prevent moonlight
from being extended and becoming an standard:
- Nobody needs it.
- Due to patent and FUD pacts, you'll need to pay MS a patent tax in
order to ever distribute moonlight which would make it a terrible
standard.
Miguel de Icaza
It seems like you did not read my blog entry, I suggest you read
it again in detail.
Miguel de Icaza
You are correct in your observation of IronPython being originally
open source.
The license they are using for now is the MS-PL, which is fine.
Miguel.
Miguel de Icaza
And? Every 10 lines of code reads on a stupid patent, that barely
prevents people from writing code.
If you are scared of patents, now is a good time to leave the software
industry behind and move to a safer space. Let me suggest something
fancy like organic farming or setting up a deli.
Miguel.
jeff.s...@gmail.com
that you completely ignore already when you load up mplayer or xine?
Anyone that wants to embrace/extend Moonlight can easily do so - they
can either license the codecs from MPEG-LA or they can drop those
codecs and go with Ogg/Theora.
There really is nothing stopping anyone from embracing and extending
Moonlight.
Anyone who plays the patent fear mongering card knows he has already
lost the argument - there's not a single piece of useful software out
there that is safe from patents.
jeff.s...@gmail.com
> Regarding your last question, these are things that prevent moonlight
> from being extended and becoming an standard:
>
> - Nobody needs it.
> - Due to patent and FUD pacts, you'll need to pay MS a patent tax in
> order to ever distribute moonlight which would make it a terrible
> standard.
"Nobody needs it": Well, if that's how you want to argue... nobody /
needs/ anything beyond food, water and air to breath. People do /want/
it, though.
Moonlight: 1 Vexorian: 0
Patents: No software is safe from patents. There's nothing
fundamentally new in Silverlight that you couldn't use prior art to
refute. The only patents you'd have to worry about are the ones for
the audio/video codecs, but that can be remedied in two ways: 1.
license the codecs like everyone else (including Microsoft) has to do;
2. use Ogg/Theora instead.
Moonlight: 2 Vexorian: 0
Thanks for playing... better luck next time.
sghe...@hotmail.com
It *is* important to see change, especially in a bad ('sub-optimal')
relationship, or you will be for-ever sour and causing your own
pain :) This goes for marriage, school, work, and communities, as it
happens.
The 'lagging' public opinion that causes people to miss their golden
opportunities. The goal is not to get the 'anti-social' kid out of the
playground, but to get everyone to play nice !
Cheers
Dan
two key concerns:
1) The VC-1 codec (a major point of Rob's that didn't get any mention
here).
2) The point that Microsoft may, at any future point, be less
cooperative than they are now. Rob's point wasn't that they aren't
cooperative now, it was that very little stops them from doing a 180
in the future (presumably once Microsoft has the upper hand in the RIA
market). Standards, when done properly, help discourage this behavior.
For example, they can be used in legal settings to establishing
evidence of anti-competitive practices (see the current Opera/
Microsoft case in the EU).
Now, this was a good point, IMO:
"You believe that without Moonlight they would not have a chance of
success, and I believe that they would have regardless of us."
This is why I don't lose much sleep over Moonlight.
This wasn't such a good point, IMO:
"You talk about Microsoft's control over Silverlight. What prevents
anyone from taking the Moonlight source code, embracing it...turning
concern. What good is a fork of Moonlight or a standardization of
Moonlight if most of the developers in the world are already targeting
an incompatible, de facto standard that can maliciously deviate at the
whim of a single company? That is the concern you need to address.
Dan
Miguel de Icaza
> here).
Am not sure that I can add anything worth to the codec discussion
beyond what has already been hashed to death on the places like the
what-wg mailing list.
Robert seems to be fine with proprietary codecs like H.264 because
they are a standard, but VC-1 and H.264 are both patent encumbered.
Luckily, Moonlight supports Ogg/Vorbis/Theora.
> 2) The point that Microsoft may, at any future point, be less
> cooperative than they are now. Rob's point wasn't that they aren't
> cooperative now, it was that very little stops them from doing a 180
> in the future (presumably once Microsoft has the upper hand in the RIA
> market). Standards, when done properly, help discourage this behavior.
I guess I fail to see the point in preemptively freak out about one of
the million things that could happen in the future.
You could play a million what-if scenarios, but if this is the one
that interests you, here is my what-if prediction based on the above
assumption: if it ever happens that they get the upper hand in the RIA
space and they no longer want to work together with us, then we can
continue doing this work alone.
The same is currently true of Mono, we did not need their cooperation
or endorsement to go ahead with it; Neither did Samba, neither did
OpenOffice, and neither does the rest of open source.
It is always better to work together for the benefit of end users, but
if it is not possible, we are not going to jump off a cliff for doing
that.
> For example, they can be used in legal settings to establishing
> evidence of anti-competitive practices (see the current Opera/
> Microsoft case in the EU).
Well, I considered Opera going to the EU a joke. If Firefox could
get 30% of the market, why cant they?
So am not sure that debating things on the light of that has any merits.
> "You talk about Microsoft's control over Silverlight. What prevents
> anyone from taking the Moonlight source code, embracing it...turning
> it into a standard? Nothing."
>
> This is true, but it makes a point that doesn't directly address the
> concern. What good is a fork of Moonlight or a standardization of
> Moonlight if most of the developers in the world are already targeting
> an incompatible, de facto standard that can maliciously deviate at the
> whim of a single company? That is the concern you need to address.
That comment addresses the issue of *control*. The claim is made
that *only* Microsoft can control the destiny of Silverlight, and that
is just showing a lack of imagination (either that, or you are trying
to make a point out of thin air, see the link about conspiracy
theorists on my blog entry).
Now, lets work with your worst case scenario: Microsoft no longer
collaborates with us, and changes Silverlight, and convinces
programmers to use the new features that they have not helped us
develop. Well, what is the worst case scenario Daniel? Do I really
have to spell it out for you?
I guess I do. Well, Microsoft will still have to publish the new
APIs so that developers use the new APIs. With no docs, there are no
developers using the new features. It is very simple. People cant
just "stumble" by accident into an unsupported API.
And guess what? When the API is documented, *gasp* we can implement it!
In fact, before our collaboration of September 4th with Microsoft we
had implemented most of Silverlight *without* their help. I know, we
are too awesome.
It seems like you are too afraid of living.
Miguel.
jakeon...@gmail.com
I couldn't agree more with your comments. I'm not (yet) a fan of
silverlight (I think it has a ways to go before it can best Flex) but
I think your approach toward silverlight and Microsoft is the right
one.
-Jake
Miguel de Icaza
Am on the same page as you.
And I think every Silverlight developer can not wait for the Mix 08
conference to see what there is in store for us.
matthew.jos...@gmail.com
a technology but as a way for case study for MS/OSS to learn to
successfully cooperate with an OSS project and provide (tacit) support
for competing desktop platforms. It will be interesting to see if MS
modify's its proprietary mindset whose mantra is "total control" and
realises that its ends can be served by cooperating with a community
of like minded software projects, potentially gaining a share of a
vastly bigger pie than MS could have created on its own.
But equally, its not all about MS modifying its mindset. If things go
well, MS continues to share specifications with Miguel and Silverlight/
Moonlight coexist and remain compliant with each other, then i will be
glad to see much of the orthodox regarding MS disappear to the annals
of software history. A tenet of OSS is cooperation, OSS needs to
strike up more cooperation with MS, we need to see more attempts to
reach out and to help proprietary competitors/shareholders realise its
never an all or nothing equation. Hopefully this iwill happen.
Its refreshing to hear the opinion of a romantic, it offers so much
more hope than those so defensive as to fear cooperation with MS.
jeff.s...@gmail.com
> Its refreshing to hear the opinion of a romantic, it offers so much
> more hope than those so defensive as to fear cooperation with MS.
andrew....@gmail.com
wrote:
> > silverlight (I think it has a ways to go before it can best Flex) but
> > I think your approach toward silverlight and Microsoft is the right
> > one.
>
> Am on the same page as you.
reasons I stand behind OSS in general. "It's worth fighting for", or
something equally as grandiose. ;)
I'll be sure to share it!
Cheers,
Andrew
Robert O'Callahan
busy for that, so I'll try to keep my response focused. I addressed
some of you rminor points in my blog comments. Basically, though, I
still don't see the strategy. Here's my best guess, tell me how
accurate it is.
"Siliverlight support will inevitably become essential for desktop
operating systems. Implementing it earlier rather than later improves
the chance Linux will have good support (and this advantage outweighs
the damage we do to open platforms by adding cross-platform
credibility to Silverlight). In the long term, we will undermine
Microsoft's control of the platform [here I'm starting to struggle] by
capturing lots of Linux market share and using that as a lever to
influence Microsoft? By deploying Moonlight on all platforms and
embracing and extending Silverlight?"
Or is it that you feel you don't need a strategy? That just writing
code is enough, and that if you write enough good code things will
work out well in the end somehow?
Robert O'Callahan
wrote:
> they are a standard, but VC-1 and H.264 are both patent encumbered.
less than happy about H.264.
> Luckily, Moonlight supports Ogg/Vorbis/Theora.
But Silverlight never will because you have no leverage to get
Microsoft to support it, so no Silverlight apps in the wild will use
it.
> > 2) The point that Microsoft may, at any future point, be less
> > cooperative than they are now. Rob's point wasn't that they aren't
> > cooperative now, it was that very little stops them from doing a 180
> > in the future (presumably once Microsoft has the upper hand in the RIA
> > market). Standards, when done properly, help discourage this behavior.
>
> I guess I fail to see the point in preemptively freak out about one of
> the million things that could happen in the future.
isn't just "one of a million things", it's what Microsoft does *all
the time*. Remember how they cooperated with CIFS documentation while
competing with Novell, then stopped when Novell was beaten? How they
cooperated with Web standards bodies while IE was the underdog vs
Netscape, then when IE won, their standards reps disappeared and their
standards support rotted?
My concerns are not fanciful speculation or paranoia, just
extrapolation from past behaviour.
> You could play a million what-if scenarios, but if this is the one
> that interests you, here is my what-if prediction based on the above
> assumption: if it ever happens that they get the upper hand in the RIA
> space and they no longer want to work together with us, then we can
> continue doing this work alone.
>
> The same is currently true of Mono, we did not need their cooperation
> or endorsement to go ahead with it; Neither did Samba, neither did
> OpenOffice, and neither does the rest of open source.
interoperability and to stay competitive, endlessly cloning
Microsoft's features and bugs and playing catch-up. They have no
choice because Microsoft already conquered those markets on its own.
On the other hand, you are helping Microsoft conquer a new market and
thereby helping *create* a playing field tilted against free software!
> I guess I do. Well, Microsoft will still have to publish the new
> APIs so that developers use the new APIs. With no docs, there are no
> developers using the new features. It is very simple. People cant
> just "stumble" by accident into an unsupported API.
>
> And guess what? When the API is documented, *gasp* we can implement it!
behaviour is unspecified in documentation and developers come to rely
on that unspecified behaviour without even realizing it, and you have
to reverse engineer that behaviour (even if it's buggy) and clone it.
Furthermore there's stuff that's not developer-facing, like protocols
to communicate with Microsoft servers and formats for data produced by
Microsoft tools, that you might also have to reverse engineer, if
Microsoft so chooses.
Moonlight will be in the same position as Samba and OpenOffice and
Mono: always behind, always having to work harder ("on the treadmill",
as Microsoft people say), always less compatible with the de facto
standard. I'm amazed that you're OK with that scenario. It doesn't
sound very romantic to me.
I am happier continuing to work on open Web technologies where the
playing field isn't tilted against us by design.
I appreciate the virtues of positive thinking and being open to
unforseen opportunities --- that's the story of Mozilla's success ---
but they can't be an excuse for being naive. You've seen the emails,
you know the history, you have to know that Microsoft's strategists
are playing for Microsoft to win and for free software to lose.
Rob
Dan
write all that :)
On Jan 4, 11:33 am, "Miguel de Icaza" <miguel.de.ic...@gmail.com>
wrote:
>
> beyond what has already been hashed to death on the places like the
> what-wg mailing list.
>
> Robert seems to be fine with proprietary codecs like H.264 because
> they are a standard, but VC-1 and H.264 are both patent encumbered.
sees VC-1 as a bigger threat than H.264?
> Luckily, Moonlight supports Ogg/Vorbis/Theora.
Yes, this is good.
> > 2) The point that Microsoft may, at any future point, be less
> > cooperative than they are now. Rob's point wasn't that they aren't
> > cooperative now, it was that very little stops them from doing a 180
> > in the future (presumably once Microsoft has the upper hand in the RIA
> > market). Standards, when done properly, help discourage this behavior.
>
> I guess I fail to see the point in preemptively freak out about one of
> the million things that could happen in the future.
can be unwise. It is wise, though, to assess the likelihood of various
events, and if that likelihood is high enough, plan one's decisions
around them.
You may "fail to see the point" because you feel that those events are
sufficiently unlikely that they can be dismissed. I'm guessing that's
where the "million things" came from, because you see the scenario as
likely as millions of other things, so why worry about it. In that
case, you differ from others because they see them as likely enough
that one should be concerned (or "freak out", as you put it).
It is this whole idea of planning for the present around probable
futures (or "preemptive freaking out") that is driving me to
investigate this debate. In the end, I will arrive at a conclusion
that will dictate a course of action. If that's freaking out, so be
it.
> You could play a million what-if scenarios, but if this is the one
> that interests you
people I know (of respectable opinion) that believe that that scenario
is likely enough to warrant consideration. That's what makes this
stand out from the other million what-ifs.
> here is my what-if prediction based on the above
> assumption: if it ever happens that they get the upper hand in the RIA
> space and they no longer want to work together with us, then we can
> continue doing this work alone.
> The same is currently true of Mono, we did not need their cooperation
> or endorsement to go ahead with it; Neither did Samba, neither did
> OpenOffice, and neither does the rest of open source.
the polar opposite, though, something akin to the game between Apple
and the iPhone devs (obviously there are technical differences).
Development is hard enough, I wouldn't want this additional battle. I
know, I know, you think there's a million-in-one chance that this
would even happen. I should just take your word for it, right?
FYI, that's where I'm coming from. I don't have any strong issues with
Mono/Silverlight. But I'm still not convinced (yet) that the approach
embodied in them is the "best" for the free software world. That
doesn't mean other people shouldn't pursue it, though.
> > For example, they can be used in legal settings to establishing
> > evidence of anti-competitive practices (see the current Opera/
> > Microsoft case in the EU).
>
> Well, I considered Opera going to the EU a joke. If Firefox could
> get 30% of the market, why cant they?
But, I'm interested in the argument that proper standards provide a
type of formal (albeit weak) agreement that can be used in various
ways, including establishing deliberate non-conformance as an anti-
competitive practice.
When you herald ECMA-334, 335, etc, what is/are _your_ "standards are
good" argument(s) (not a rhetorical question)?
> > This is true, but it makes a point that doesn't directly address the
> > concern. What good is a fork of Moonlight or a standardization of
> > Moonlight if most of the developers in the world are already targeting
> > an incompatible, de facto standard that can maliciously deviate at the
> > whim of a single company? That is the concern you need to address.
>
> That comment addresses the issue of *control*. The claim is made
> that *only* Microsoft can control the destiny of Silverlight, and that
> is just showing a lack of imagination (either that, or you are trying
> to make a point out of thin air, see the link about conspiracy
> theorists on my blog entry).
imaginative way that a non-Microsoft entity can control (not just
influence) the destiny of Silverlight. No credit for an unlikely,
million-to-one what-if scenario...even if you freak out :)
> Now, lets work with your worst case scenario: Microsoft no longer
> collaborates with us, and changes Silverlight, and convinces
> programmers to use the new features that they have not helped us
> develop. Well, what is the worst case scenario Daniel? Do I really
> have to spell it out for you?
> I guess I do. Well, Microsoft will still have to publish the new
> APIs so that developers use the new APIs. With no docs, there are no
> developers using the new features.
> And guess what? When the API is documented, *gasp* we can implement it!
iPhone, in that there is an API meant for public consumption (APIs for
privileged consumption can be dealt with, too). That does make it
easier. There are issues like undocumented _behavior_ that also throw
in a wrench. But, obviously the WINE guys get by, right?
But, is this the life I want to live? When other choices are placed
before me, especially _before_ a particular technology takes hold,
should I not consider the potential pitfalls of each approach?
> In fact, before our collaboration of September 4th with Microsoft we
> had implemented most of Silverlight *without* their help. I know, we
> are too awesome.
Moonlight built upon previous work of thousands of developers. Heck,
even I had a small part in it:
http://www.google.com/search?q=site%3Alists.freedesktop.org+amelang
http://gitweb.freedesktop.org/?p=cairo&a=search&h=HEAD&s=amelang
(FYI, the Win32-related commits don't mean I run Windows, in case
you're looking for ammo for a personal attack)
> It seems like you are too afraid of living.
but not against me. Do you consider my single comment sufficient to
arrive at this conclusion?
After reading much of what you have written over the years (code and
blog), I have reserved reaching any deep judgment about your
character. Please return the favor.
Dan
Dan
>
> Its refreshing to hear the opinion of a romantic, it offers so much
> more hope than those so defensive as to fear cooperation with MS.
for in an argument, we might as well sign up here, right :)
http://www.venganza.org/
So hope for the sake of hope isn't enough for me. That's why I'm
asking so many questions. I want to place my hope carefully.
Dan
Miguel de Icaza
> > they are a standard, but VC-1 and H.264 are both patent encumbered.
>
> Sure, that's Robert's opinion, not mine. Do you understand why Robert
> sees VC-1 as a bigger threat than H.264?
I do not. They are both patent encumbered, from a FOSS standpoint,
they are both bad.
You seem to be trying to convince an Afghani that the bombing of his
country by the americans is better than being bombed by the soviets.
> It interests me only in that there is a sufficient enough number of
> people I know (of respectable opinion) that believe that that scenario
> is likely enough to warrant consideration. That's what makes this
> stand out from the other million what-ifs.
Its called fear peddling. It works great to advance any agenda. See US, 2003.
> I agree, you don't need cooperation nor endorsement. You can also have
> the polar opposite, though, something akin to the game between Apple
> and the iPhone devs (obviously there are technical differences).
Yeah, that would be annoying.
Luckily that is not the current situation, and hopefully we can find a
win-win situation for Microsoft and the open sourcistas. So am
planning on doing everything to help Microsoft become a better
netizen.
> When you herald ECMA-334, 335, etc, what is/are _your_ "standards are
> good" argument(s) (not a rhetorical question)?
I like those because they published information that was not easily
obtainable from sites like MSDN. The value for stuff that is in MSDN
to become part of the standard is less obvious (like class libraries).
> I have been known to have lapses of imagination, so humor me with an
> imaginative way that a non-Microsoft entity can control (not just
> influence) the destiny of Silverlight. No credit for an unlikely,
> million-to-one what-if scenario...even if you freak out :)
Here is a simple example: you take Moonlight and fork it to include a
useful method "createFromJson" in addition to the "createFromXaml"
that it has. Distribute it. Repeat for any cute innovation that you
want until you are tired.
As for us, we are going to stick to be compatible, because we value
two-way compatibility. But if the time comes when Microsoft is not
innovating with Silverilght and they abandon it, we can take things
into our own hands.
> That does make it
> easier. There are issues like undocumented _behavior_ that also throw
> in a wrench. But, obviously the WINE guys get by, right?
Its not a big deal, and its not worse than differences across web
browsers for things that are already fully spelled out, like CSS.
Do not drown yourself in a glass of water.
> You may have some basis for your ad hominem arguments against Robert,
> but not against me. Do you consider my single comment sufficient to
> arrive at this conclusion?
You are confusing ad hominem attack by observed behavior.
Miguel
Miguel de Icaza
> less than happy about H.264.
This was related to this comment of yours: "The codec situation is
also better: H264 is patent encumbered but at least it's a standard
and it's not controlled by one company."
Maybe I should not have said that you are "ok" with it, but you seem
to prefer it.
> But Silverlight never will because you have no leverage to get
> Microsoft to support it, so no Silverlight apps in the wild will use
> it.
At least we are actively contributing to its adoption.
Maybe -we- Linux users are a small user base, but on Linux we will be
bale to use free codecs for our desktop needs.
But I get the feeling that what you are trying to imply is that only
Mozilla can make "ogg" universal, is that the case?
> Argh! "Leveraging developer platform dominance to maintain monopoly"
> isn't just "one of a million things", it's what Microsoft does *all
> the time*.
And you *seriously* believe had Moonlight not existed would have
stopped Silverlight? If you do, we should talk, because I got a
bridge to sell you.
> Remember how they cooperated with CIFS documentation while
> competing with Novell, then stopped when Novell was beaten?
Am not that old, I do not know that story.
> How they
> cooperated with Web standards bodies while IE was the underdog vs
> Netscape, then when IE won, their standards reps disappeared and their
> standards support rotted?
That I remember; But I also remember "High Stakes, No Prisoners" a
story of incompetence on the part of Microsoft's competitors (a book
written by someone that despised Microsoft).
> My concerns are not fanciful speculation or paranoia, just
> extrapolation from past behaviour.
Except that things change. You are not willing to give change a chance.
There was a time when Sun would not consider open sourcing Java.
Look where we are today.
> On the other hand, you are helping Microsoft conquer a new market and
> thereby helping *create* a playing field tilted against free software!
Well, thanks for all the credit that you give us. But I disagree
with your conclusion, and also, you do not seem to address any of the
core issues: you are not delivering something that developers want.
> You know as well as I do that this is not the whole story. A lot of
> behaviour is unspecified in documentation and developers come to rely
> on that unspecified behaviour without even realizing it, and you have
> to reverse engineer that behaviour (even if it's buggy) and clone it.
You are correct that there is more to the story, but the real weight
of these hidden problems is very small. We have been doing this
since 2001, and the ratio of undocumented/side effect issues to the
documentation is negligible. The majority of our bugs are genuine
bugs, half-implemented features, stubbed code, not unknown really side
effects.
> Furthermore there's stuff that's not developer-facing, like protocols
> to communicate with Microsoft servers and formats for data produced by
> Microsoft tools, that you might also have to reverse engineer, if
> Microsoft so chooses.
Sure, and the same can happen anywhere else. They could add special
HTML controls that only work with proprietary servers or proprietary
extensions that only work with their engine, and like anyone else we
would have to evaluate if its worth cloning or not. See
XmlHttpRequest.
> Moonlight will be in the same position as Samba and OpenOffice and
> Mono: always behind, always having to work harder ("on the treadmill",
> as Microsoft people say), always less compatible with the de facto
> standard. I'm amazed that you're OK with that scenario. It doesn't
> sound very romantic to me.
That is because you are being defeatist.
I used to be a sysadmin before starting Ximian, and as a Samba
sysadmin, Samba could do *so much more* than you could do with Windows
and Lan Manager that it is not even funny. There is just no
comparison!
OpenOffice being free-for-all is a feature that you can not compete
against, and if you think that merely copying Office has been all OOo
can do, you do not know much about OpenOffice. Didnt OOo lead the
way in XML file formats? (Well, they didnt, but they were the first
ones with the volume to do so).
With no OpenOffice, there would be no ODF.
> I am happier continuing to work on open Web technologies where the
> playing field isn't tilted against us by design.
Am glad that you are, am a happy user of your work.
> I appreciate the virtues of positive thinking and being open to
> unforseen opportunities --- that's the story of Mozilla's success ---
> but they can't be an excuse for being naive. You've seen the emails,
> you know the history, you have to know that Microsoft's strategists
> are playing for Microsoft to win and for free software to lose.
The emails are not different than the emails that you will find in any
other organization. Microsoft had to go through the public
embarrassment of having to disclose their records due to their
lawsuit. But I bet you they are no different in any other company
that you can get subpoenaed.
As for change, I believe there are many winds of change inside
Microsoft. Starting with people in Scott Guthrie's group that
understands the value of openness. I have blogged about this
previously. Microsoft is not going to become the next Sun when it
comes to opening up code, but am happy with taking steps in the right
direction.
Miguel.
Miguel de Icaza
> I don't want to get into a tight loop discussion this, we're both too
> busy for that, so I'll try to keep my response focused. I addressed
> some of you rminor points in my blog comments. Basically, though, I
> still don't see the strategy. Here's my best guess, tell me how
> accurate it is.
Let me summarize it quickly:
* Need Silverlight on Linux to keep Silverlight as a second class citizen.
* Silverlight has features that nothing else provides today.
* I will take collaboration over confrontation any day.
As long as Microsoft continues to innovate on Silverlight and improve
in the same way/direction/taste than C# I see no reason to fork,
specially with our limited resources. But if Microsoft would to
severely botch Silverlight, abandon it or something similar there
would be a case in my mind to fork and deviate from their feature set.
In addition, we are *today* using Moonlight in the desktop (we have no
reason to limit it on the web) which is just an extra bonus,
Miguel.
Dan
wrote:
>
> country by the americans is better than being bombed by the soviets.
this thread, regardless of how it "seems" to you (feel free to scroll
up). How "convincing" can I be? :) I came here looking for the other
side of an argument (that wasn't even mine!).
> > It interests me only in that there is a sufficient enough number of
> > people I know (of respectable opinion) that believe that that scenario
> > is likely enough to warrant consideration. That's what makes this
> > stand out from the other million what-ifs.
>
> Its called fear peddling. It works great to advance any agenda. See US, 2003.
doesn't qualify for fear peddling, right? So, I don't follow the leap
here. At what point does it become fear mongering for you?
> Luckily that is not the current situation, and hopefully we can find a
> win-win situation for Microsoft and the open sourcistas. So am
> planning on doing everything to help Microsoft become a better
> netizen.
even probable. I haven't made the leap yet to actually bet on this. I
call this caution, I know, you think it's fear.
> > When you herald ECMA-334, 335, etc, what is/are _your_ "standards are
> > good" argument(s) (not a rhetorical question)?
>
> I like those because they published information that was not easily
> obtainable from sites like MSDN. The value for stuff that is in MSDN
> to become part of the standard is less obvious (like class libraries).
That's fine. This part does explain a lot about the different of
opinions in this discussion, which is good.
> > I have been known to have lapses of imagination, so humor me with an
> > imaginative way that a non-Microsoft entity can control (not just
> > influence) the destiny of Silverlight. No credit for an unlikely,
> > million-to-one what-if scenario...even if you freak out :)
>
> Here is a simple example: you take Moonlight and fork it to include a
> useful method "createFromJson" in addition to the "createFromXaml"
> that it has. Distribute it. Repeat for any cute innovation that you
> want until you are tired.
merely influencing. In other words, I don't think there is enough
"bite" in the above approach to "control the destiny of Silverlight".
But, whatever, we're both speculating, let's just see what happens
over time.
> Do not drown yourself in a glass of water.
exactly how much water there is.
> > You may have some basis for your ad hominem arguments against Robert,
> > but not against me. Do you consider my single comment sufficient to
> > arrive at this conclusion?
>
> You are confusing ad hominem attack by observed behavior.
based on observed behavior. In fact, the attack could be based
entirely on well-established facts. But it remains a formal fallacy
which intentionally or unintentionally can turn the focus away from
the main topic onto an individual. It can give the illusion of making
a point against or for an argument, but in fact it only distracts from
the discussion by shaming a participant.
If you revisit each one of your contributions today (in Rob's blog's
comments, your blog entry, and the comments here), you'll find at
least four cases where you veer off and make a negative jab at an
individual/organization in addition to, or rather than making a valid
point.
Consider how you brushed off my point about the Opera/Microsoft case.
You said that case was a joke, then asked why Opera can't get market
share comparable to Firefox, and then said, "So am not sure that
debating things on the light of that has any merits." As if the debate
lost merit because Opera sucks? The point I was making only
tangentially involved Opera, how could the whole thing be shot?
Anyway, I'm sure we both need to get back to coding free software, so
regardless of how well you put me in my place in your reply, I promise
to resist responding :) I'll read it of course.
Best wishes on your endeavors, at least our end game is the same, and
who knows, if your plan succeeds, I may be hacking on Mono/Moonlight
one day :)
Dan
Robert O'Callahan
wrote:
> Mozilla can make "ogg" universal, is that the case?
digression really...
> > Argh! "Leveraging developer platform dominance to maintain monopoly"
> > isn't just "one of a million things", it's what Microsoft does *all
> > the time*.
>
> And you *seriously* believe had Moonlight not existed would have
> stopped Silverlight?
its adoption might be slower or more limited without Moonlight, based
on comments by people who say "ah, but is it cross-platform including
Linux?" and get the answer "Moonlight!".
> > Remember how they cooperated with CIFS documentation while
> > competing with Novell, then stopped when Novell was beaten?
>
> Am not that old, I do not know that story.
an archive of everything Microsoft's done since the early 80s, it
would be worth your study.
> > How they
> > cooperated with Web standards bodies while IE was the underdog vs
> > Netscape, then when IE won, their standards reps disappeared and their
> > standards support rotted?
>
> That I remember; But I also remember "High Stakes, No Prisoners" a
> story of incompetence on the part of Microsoft's competitors (a book
> written by someone that despised Microsoft).
behaviour patterns here.
> > My concerns are not fanciful speculation or paranoia, just
> > extrapolation from past behaviour.
>
> Except that things change. You are not willing to give change a chance.
>
> There was a time when Sun would not consider open sourcing Java.
> Look where we are today.
fundamentally changed. That's interesting.
> > On the other hand, you are helping Microsoft conquer a new market and
> > thereby helping *create* a playing field tilted against free software!
>
> Well, thanks for all the credit that you give us. But I disagree
> with your conclusion, and also, you do not seem to address any of the
> core issues: you are not delivering something that developers want.
seems. Maybe there are other approaches you could have tried to
satisfy other developers. For example, what about pitching in to help
make Mono/C# a first-class Web scripting language?
> > You know as well as I do that this is not the whole story. A lot of
> > behaviour is unspecified in documentation and developers come to rely
> > on that unspecified behaviour without even realizing it, and you have
> > to reverse engineer that behaviour (even if it's buggy) and clone it.
>
> You are correct that there is more to the story, but the real weight
> of these hidden problems is very small. We have been doing this
> since 2001, and the ratio of undocumented/side effect issues to the
> documentation is negligible. The majority of our bugs are genuine
> bugs, half-implemented features, stubbed code, not unknown really side
> effects.
that. I wonder where Silverlight will land.
> > Furthermore there's stuff that's not developer-facing, like protocols
> > to communicate with Microsoft servers and formats for data produced by
> > Microsoft tools, that you might also have to reverse engineer, if
> > Microsoft so chooses.
>
> Sure, and the same can happen anywhere else. They could add special
> HTML controls that only work with proprietary servers or proprietary
> extensions that only work with their engine, and like anyone else we
> would have to evaluate if its worth cloning or not. See
> XmlHttpRequest.
developers that we don't implement that, or they should do it
differently in a standards-based way. Whenever Microsoft adds
something to Silverlight that's automatically going to be part of the
(de facto) standard.
> > Moonlight will be in the same position as Samba and OpenOffice and
> > Mono: always behind, always having to work harder ("on the treadmill",
> > as Microsoft people say), always less compatible with the de facto
> > standard. I'm amazed that you're OK with that scenario. It doesn't
> > sound very romantic to me.
>
> That is because you are being defeatist.
>
> I used to be a sysadmin before starting Ximian, and as a Samba
> sysadmin, Samba could do *so much more* than you could do with Windows
> and Lan Manager that it is not even funny. There is just no
> comparison!
and actually they were doing just fine?
> OpenOffice being free-for-all is a feature that you can not compete
> against, and if you think that merely copying Office has been all OOo
> can do, you do not know much about OpenOffice. Didnt OOo lead the
> way in XML file formats? (Well, they didnt, but they were the first
> ones with the volume to do so).
>
> With no OpenOffice, there would be no ODF.
accomplished in spite of the tilted playing field, but is MSOffice
compatibility a done deal now? (Is Microsoft collaborating with Novell
on the OOXML format for the next version of MSOffice so that
OpenOffice will be able to handle it from the beginning? Are they
making changes to ensure that OpenOffice will be able to implement it
without unnecessary pain? They would be if it was a truly multi-vendor
standard. Brr, I probably shouldn't open this can of worms!)
> Am glad that you are, am a happy user of your work.
> > I appreciate the virtues of positive thinking and being open to
> > unforseen opportunities --- that's the story of Mozilla's success ---
> > but they can't be an excuse for being naive. You've seen the emails,
> > you know the history, you have to know that Microsoft's strategists
> > are playing for Microsoft to win and for free software to lose.
>
> The emails are not different than the emails that you will find in any
> other organization. Microsoft had to go through the public
> embarrassment of having to disclose their records due to their
> lawsuit. But I bet you they are no different in any other company
> that you can get subpoenaed.
like I said in my blog, *no* single company can be trusted (although
if their code is free software, the risks are a lot lower).
Having said that, I think it's sensible to be more cautious about the
heavily-armed bully than the lightly-armed one, so I worry more about
Microsoft with its portfolio of monopolies and cast-iron revenue
streams than, say, Adobe.
> As for change, I believe there are many winds of change inside
> Microsoft. Starting with people in Scott Guthrie's group that
> understands the value of openness. I have blogged about this
> previously. Microsoft is not going to become the next Sun when it
> comes to opening up code, but am happy with taking steps in the right
> direction.
others are in charge, I don't expect to see major changes in the way
they compete. (Except as they bureaucratize and become generally less
effective.)
Rob
Robert O'Callahan
wrote:
> Mozilla can make "ogg" universal, is that the case?
digression really...
> > isn't just "one of a million things", it's what Microsoft does *all
> > the time*.
>
> And you *seriously* believe had Moonlight not existed would have
> stopped Silverlight?
its adoption might be slower or more limited without Moonlight, based
on comments by people who say "ah, but is it cross-platform including
Linux?" and get the answer "Moonlight!".
> > competing with Novell, then stopped when Novell was beaten?
>
> Am not that old, I do not know that story.
an archive of everything Microsoft's done since the early 80s, it
would be worth your study.
> > cooperated with Web standards bodies while IE was the underdog vs
> > Netscape, then when IE won, their standards reps disappeared and their
> > standards support rotted?
>
> That I remember; But I also remember "High Stakes, No Prisoners" a
> story of incompetence on the part of Microsoft's competitors (a book
> written by someone that despised Microsoft).
behaviour patterns here.
> > extrapolation from past behaviour.
>
> Except that things change. You are not willing to give change a chance.
>
> There was a time when Sun would not consider open sourcing Java.
> Look where we are today.
fundamentally changed. That's interesting.
> > thereby helping *create* a playing field tilted against free software!
>
> Well, thanks for all the credit that you give us. But I disagree
> with your conclusion, and also, you do not seem to address any of the
> core issues: you are not delivering something that developers want.
seems. Maybe there are other approaches you could have tried to
satisfy other developers. For example, what about pitching in to help
make Mono/C# a first-class Web scripting language?
> > behaviour is unspecified in documentation and developers come to rely
> > on that unspecified behaviour without even realizing it, and you have
> > to reverse engineer that behaviour (even if it's buggy) and clone it.
>
> You are correct that there is more to the story, but the real weight
> of these hidden problems is very small. We have been doing this
> since 2001, and the ratio of undocumented/side effect issues to the
> documentation is negligible. The majority of our bugs are genuine
> bugs, half-implemented features, stubbed code, not unknown really side
> effects.
that. I wonder where Silverlight will land.
> > to communicate with Microsoft servers and formats for data produced by
> > Microsoft tools, that you might also have to reverse engineer, if
> > Microsoft so chooses.
>
> Sure, and the same can happen anywhere else. They could add special
> HTML controls that only work with proprietary servers or proprietary
> extensions that only work with their engine, and like anyone else we
> would have to evaluate if its worth cloning or not. See
> XmlHttpRequest.
developers that we don't implement that, or they should do it
differently in a standards-based way. Whenever Microsoft adds
something to Silverlight that's automatically going to be part of the
(de facto) standard.
> > Mono: always behind, always having to work harder ("on the treadmill",
> > as Microsoft people say), always less compatible with the de facto
> > standard. I'm amazed that you're OK with that scenario. It doesn't
> > sound very romantic to me.
>
> That is because you are being defeatist.
>
> I used to be a sysadmin before starting Ximian, and as a Samba
> sysadmin, Samba could do *so much more* than you could do with Windows
> and Lan Manager that it is not even funny. There is just no
> comparison!
and actually they were doing just fine?
> against, and if you think that merely copying Office has been all OOo
> can do, you do not know much about OpenOffice. Didnt OOo lead the
> way in XML file formats? (Well, they didnt, but they were the first
> ones with the volume to do so).
>
> With no OpenOffice, there would be no ODF.
accomplished in spite of the tilted playing field, but is MSOffice
compatibility a done deal now? (Is Microsoft collaborating with Novell
on the OOXML format for the next version of MSOffice so that
OpenOffice will be able to handle it from the beginning? Are they
making changes to ensure that OpenOffice will be able to implement it
without unnecessary pain? They would be if it was a truly multi-vendor
standard. Brr, I probably shouldn't open this can of worms!)
> > I appreciate the virtues of positive thinking and being open to
> > unforseen opportunities --- that's the story of Mozilla's success ---
> > but they can't be an excuse for being naive. You've seen the emails,
> > you know the history, you have to know that Microsoft's strategists
> > are playing for Microsoft to win and for free software to lose.
>
> The emails are not different than the emails that you will find in any
> other organization. Microsoft had to go through the public
> embarrassment of having to disclose their records due to their
> lawsuit. But I bet you they are no different in any other company
> that you can get subpoenaed.
like I said in my blog, *no* single company can be trusted (although
if their code is free software, the risks are a lot lower).
Having said that, I think it's sensible to be more cautious about the
heavily-armed bully than the lightly-armed one, so I worry more about
Microsoft with its portfolio of monopolies and cast-iron revenue
streams than, say, Adobe.
> Microsoft. Starting with people in Scott Guthrie's group that
> understands the value of openness. I have blogged about this
> previously. Microsoft is not going to become the next Sun when it
> comes to opening up code, but am happy with taking steps in the right
> direction.
Robert O'Callahan
wrote:
> Mozilla can make "ogg" universal, is that the case?
digression really...
> > isn't just "one of a million things", it's what Microsoft does *all
> > the time*.
>
> And you *seriously* believe had Moonlight not existed would have
> stopped Silverlight?
its adoption might be slower or more limited without Moonlight, based
on comments by people who say "ah, but is it cross-platform including
Linux?" and get the answer "Moonlight!".
> > competing with Novell, then stopped when Novell was beaten?
>
> Am not that old, I do not know that story.
an archive of everything Microsoft's done since the early 80s, it
would be worth your study.
> > cooperated with Web standards bodies while IE was the underdog vs
> > Netscape, then when IE won, their standards reps disappeared and their
> > standards support rotted?
>
> That I remember; But I also remember "High Stakes, No Prisoners" a
> story of incompetence on the part of Microsoft's competitors (a book
> written by someone that despised Microsoft).
behaviour patterns here.
> > extrapolation from past behaviour.
>
> Except that things change. You are not willing to give change a chance.
>
> There was a time when Sun would not consider open sourcing Java.
> Look where we are today.
fundamentally changed. That's interesting.
> > thereby helping *create* a playing field tilted against free software!
>
> Well, thanks for all the credit that you give us. But I disagree
> with your conclusion, and also, you do not seem to address any of the
> core issues: you are not delivering something that developers want.
seems. Maybe there are other approaches you could have tried to
satisfy other developers. For example, what about pitching in to help
make Mono/C# a first-class Web scripting language?
> > behaviour is unspecified in documentation and developers come to rely
> > on that unspecified behaviour without even realizing it, and you have
> > to reverse engineer that behaviour (even if it's buggy) and clone it.
>
> You are correct that there is more to the story, but the real weight
> of these hidden problems is very small. We have been doing this
> since 2001, and the ratio of undocumented/side effect issues to the
> documentation is negligible. The majority of our bugs are genuine
> bugs, half-implemented features, stubbed code, not unknown really side
> effects.
that. I wonder where Silverlight will land.
> > to communicate with Microsoft servers and formats for data produced by
> > Microsoft tools, that you might also have to reverse engineer, if
> > Microsoft so chooses.
>
> Sure, and the same can happen anywhere else. They could add special
> HTML controls that only work with proprietary servers or proprietary
> extensions that only work with their engine, and like anyone else we
> would have to evaluate if its worth cloning or not. See
> XmlHttpRequest.
developers that we don't implement that, or they should do it
differently in a standards-based way. Whenever Microsoft adds
something to Silverlight that's automatically going to be part of the
(de facto) standard.
> > Mono: always behind, always having to work harder ("on the treadmill",
> > as Microsoft people say), always less compatible with the de facto
> > standard. I'm amazed that you're OK with that scenario. It doesn't
> > sound very romantic to me.
>
> That is because you are being defeatist.
>
> I used to be a sysadmin before starting Ximian, and as a Samba
> sysadmin, Samba could do *so much more* than you could do with Windows
> and Lan Manager that it is not even funny. There is just no
> comparison!
and actually they were doing just fine?
> against, and if you think that merely copying Office has been all OOo
> can do, you do not know much about OpenOffice. Didnt OOo lead the
> way in XML file formats? (Well, they didnt, but they were the first
> ones with the volume to do so).
>
> With no OpenOffice, there would be no ODF.
accomplished in spite of the tilted playing field, but is MSOffice
compatibility a done deal now? (Is Microsoft collaborating with Novell
on the OOXML format for the next version of MSOffice so that
OpenOffice will be able to handle it from the beginning? Are they
making changes to ensure that OpenOffice will be able to implement it
without unnecessary pain? They would be if it was a truly multi-vendor
standard. Brr, I probably shouldn't open this can of worms!)
> > I appreciate the virtues of positive thinking and being open to
> > unforseen opportunities --- that's the story of Mozilla's success ---
> > but they can't be an excuse for being naive. You've seen the emails,
> > you know the history, you have to know that Microsoft's strategists
> > are playing for Microsoft to win and for free software to lose.
>
> The emails are not different than the emails that you will find in any
> other organization. Microsoft had to go through the public
> embarrassment of having to disclose their records due to their
> lawsuit. But I bet you they are no different in any other company
> that you can get subpoenaed.
like I said in my blog, *no* single company can be trusted (although
if their code is free software, the risks are a lot lower).
Having said that, I think it's sensible to be more cautious about the
heavily-armed bully than the lightly-armed one, so I worry more about
Microsoft with its portfolio of monopolies and cast-iron revenue
streams than, say, Adobe.
> Microsoft. Starting with people in Scott Guthrie's group that
> understands the value of openness. I have blogged about this
> previously. Microsoft is not going to become the next Sun when it
> comes to opening up code, but am happy with taking steps in the right
> direction.
893prod...@gmail.com
kirk...@gmail.com
Silverlight. If it should become a replacement for Flash, I can only
see this as a HUGE step for web experiences on Linux as it will
already be implemented . But don't listen to me, I also approve of C#.
Keep up the good work.
greg
Users want:
- A canvas to write cross-platform desktop apps that integrate well
with mono.
- A browser based player for animations, and RIA's which need a
modern VM.
Strategy 1) Don't implement anything and continue to use proprietary
Flash.
- It's proprietry, not a standard and controlled by a single large
company.
- Mono developers don't get a canvas.
+ Microsoft can't play the cross platform card.
Strategy 2) Implement Silverlight.
+ Moonlight is open-source.
+ Mono users get a canvas.
+ Users can view Silverlight content on all platforms (must ffmpeg
for VC-1 on non-standard os's or architectures).
- Microsoft can play the cross platform card.
+ Windows developers can easily port and transition to linux.
Strategy 3) Implement Otherlight(TM) - i.e. implement something
similar to Silverlight, but different.
+ Otherlight is open-source, mono team can write their own standard.
- Take a long time and resource to design this type of software from
scratch.
- Since Silverlight design is good, which bits should be changed so
that its not the same?
- Windows developers can't easily port their code and transition to
linux.
+ Microsoft can't play the cross platform card.
Alternate stategies probably not worth considering:
4) Fix SVG
- I remember Roc complaining that vendors who don't care about the
web
have hijacked this standard, making it very difficult to implement
(Irony?).
5) Implement an open-source Flash player and integrate it with mono.
- Lots of legacy cruft, and probably not a good fit for mono.
As I interpret so far - please correct if I'm wrong:
Roc argues that users of open source platfroms should be willing to
take the pain of not being able to view Silverlight content, and mono
developers the pain of not having a canvas, and that this sacrifice
will make a difference to the future of the web.
Miguel argues that the sacrifice would make little difference to
Silverlight adoption, and the pain would outweight the adoption
advantage, and would reduce software freedom for developers on open
source platforms.
So that raises two questions:
1) How much pain should users of open source platforms be willing to
take on in order to make the web a better place?
2) How much difference would the sacrifice likely make in terms of
Silverlight adoption?
In my opinion, if enough developers like Microsoft's tools and API's
they will make RIA's and animations for Silverlight. And html will end
up being used just as a backup for people with mobile devices or os's
without Silverlight. Not a nice thought - but if their tools are
better it will likely happen.
I can only see two ways to stop Silverlight adoption:
1) Write better or comparable vm/apis/tools, gain majority adoption,
then standardise.
2) Make a large majority of developers and managers believe in the
free software ideals, and to be willing to take a productivity hit by
using lower quality vm/apis/tools.
Both sound pretty hard but in my opinion 1) is far more achievable.
In Miguels shoes, with his limited resources (relative to MS, Adobe or
Bigcorp), the sensible way to start is to copy the bits that the
proprietary competitor does well - isn't that exactly what Stallman
and Linus did? Their projects worked out ok didn't they?
But if Roc and Mozilla can drum up enough support of free software in
corporate tie land - great, but I don't see that as likely just yet.
Cheers,
Greg.
Arne Claassen
Boo, Python and Ruby. And that matters to me, and matters to others."
Count me among those others. I want to write web apps and cross
platform apps in my language of choice with a platform that's not a
series of hacks glued together by libs trying to overcome platform
incompatibilities (read: js+HTML). It would be nice if instead of
vilifying Silverlight because it came from Redmond and therefore
nothing even remotely related to it could ever be good, people would
take moonlight, port to every platform and make it more ubiquitous
than even Silverlight. I.e. Silverlight is browser only, but moonlight
is not bound by this. I'd love a platform that used Moonlight+mono to
write WPF styles apps across all platforms. Keep up the good work,
Miguel and all other mono supporters.
java...@gmail.com
agree. Linux must be able to provide options that will be used, but it
is a double edged sword.
The more we help the monopoly Microsoft, the more we empower them to
continue their control. Linux is more than able on the server and
desktop to overtake Microsoft with any of the MS Operating Systems,
and I for one, do not want to help Microsoft continue to keep people
locked in. The problem and paradox is that, (1) Linux must be able to
support proprietary software including MS, and (2) Supporting anything
from MS works ultimately, against us.
NET and MONO are good examples. I personally think embracing NET was
and is a very bad thing for Linux. The more we support MS' Java, the
more we empower MS to control. MONO cannot and never will be where MS'
NET is. Why? MS totally controls NET, its future, direction, features,
etc. They have recently released their source for viewing no doubt
hoping they can use this to promote a patent issue case against MONO,
should the Open Source world start actually embracing MONO and begin
to actually start building a software stack with it. Silverlight is
giving part of this ability.
But lets suppose Icaza and the MONO group WERE able to keep MONO free
from patent infringements (which will ultimately proved increasingly
difficult). As MONO conitnues in its race for features and
compatibility, MS will constantly morph NET into something else.
Ultimately, MS could pull the plug on non MS Silverlight, by changing
the NET standards on which it depends to such a degree, to make
MONO.NET's system totally unable to sustain the first iterations of
Silverlight. Once the world is dependent upon Silverlight, MONO based
Silverlight suddenly is no longer compatible. Thus, MS maintains
absolute and total control. That's the rub and thus the danger.
I would rather see a Silverlight clone based not on NET but on Java or
something else. That way regardless of what MS does, a non NET based
Silverlight can adapt without having to depend upon a NET that will
constantly change to further MS' idea of total computer and software
dominance, which is not good for the future of computing.
Some may call my view irrational, but it is the history of Microsoft.
Some believe MS has changed, they are no longer the predatory company
they were. These are the same people who see terrorist as just
misunderstood, while their heads start rolling down the hill at the
edge of the terrorists sword.
Do not fall into the trap so many companies have fell into before;
believing MS goal is peaceful co-existence. It never has been MS' goal
and it certainly is not their goal now. MS has always proved that they
cannot be trusted to do anything but kill the competition and promote
more MS dependency and lock in. MONO is a good idea, but it founded on
the premise of allowing MS to continue to exert total control. Java
and Open Source like Python, Ruby, Perl, etc. are the only acceptable
frameworks from which we can build an infrastructure that is totally
independent from commercial control.
jeff.s...@gmail.com
> I really do understand Icaza's reasoning on Silverlight and I somewhat
> agree. Linux must be able to provide options that will be used, but it
> is a double edged sword.
>
> The more we help the monopoly Microsoft, the more we empower them to
> continue their control. Linux is more than able on the server and
> desktop to overtake Microsoft with any of the MS Operating Systems,
> and I for one, do not want to help Microsoft continue to keep people
> locked in. The problem and paradox is that, (1) Linux must be able to
> support proprietary software including MS, and (2) Supporting anything
> from MS works ultimately, against us.
>
> NET and MONO are good examples. I personally think embracing NET was
> and is a very bad thing for Linux. The more we support MS' Java, the
> more we empower MS to control. MONO cannot and never will be where MS'
> NET is.
dramatically and Mono's compiler already supports most (if not all)
the latest features. I think there are bugs still, but they'll get
worked out.
if they don't update their platform with new APIs as fast as the Mono
team implements the old ones, there's no reason we can't catch up.
*Especially* if more contributors joined in and helped out ;-)
When you were in high school or university and a professor gave you a
big project to complete, perhaps your first thoughts were "wow,
there's SO much work to be done! I'll never finish it!". But you took
it one step at a time and it got done in time to be turned in, which
if your university experience was anything like mine, you weren't
given a whole lot of time.
Mono & .Net is the same thing. If you spend all your time complaining
that there's too much to do, it'll never get done. If, on the other
hand, you actually buckle down and set your mind to completing the
task, it somehow always gets done.
> Why? MS totally controls NET, its future, direction, features,
> etc. They have recently released their source for viewing no doubt
> hoping they can use this to promote a patent issue case against MONO,
> should the Open Source world start actually embracing MONO and begin
> to actually start building a software stack with it. Silverlight is
> giving part of this ability.
should we be cautious about it? yes, but we already have taken
measures to protect ourselves.
>
> But lets suppose Icaza and the MONO group WERE able to keep MONO free
> from patent infringements (which will ultimately proved increasingly
> difficult).
keep Linux free of patent infringements (which will ultimately proved
increasingly difficult)."
Does that make you feel better? :)
The same argument can be applied to any piece of software.
> As MONO conitnues in its race for features and
> compatibility, MS will constantly morph NET into something else.
> Ultimately, MS could pull the plug on non MS Silverlight, by changing
> the NET standards on which it depends to such a degree, to make
> MONO.NET's system totally unable to sustain the first iterations of
> Silverlight.
changing the .Net standards to such a degree would be suicidal to
Microsoft.
> Once the world is dependent upon Silverlight, MONO based
> Silverlight suddenly is no longer compatible. Thus, MS maintains
> absolute and total control. That's the rub and thus the danger.
which constantly adds features every year or two they release a new
version with new features that Gnash and swfdec will likely not get
around to implementing for years afterward.
But the people who bash Moonlight always seem to point to Gnash, for
example, and say "works well enough for me! I can even view YouTube!"
This suggests that the web developers are always lagging behind as far
as taking advantage of the latest and greatest feature sets. There's
no reason to believe it will be any different with Silverlight. The
difference is that Moonlight will likely always be closer to the
latest Silverlight release than Gnash will ever be to catching Flash
(unless Flash dies).
Doom and gloom? Hardly.
>
> I would rather see a Silverlight clone based not on NET but on Java or
> something else. That way regardless of what MS does, a non NET based
> Silverlight can adapt without having to depend upon a NET that will
> constantly change to further MS' idea of total computer and software
> dominance, which is not good for the future of computing.
view Silverlight content and it wouldn't take over the web either.
You really need to think before you speak :)
>
> Some may call my view irrational, but it is the history of Microsoft.
> Some believe MS has changed, they are no longer the predatory company
> they were. These are the same people who see terrorist as just
> misunderstood, while their heads start rolling down the hill at the
> edge of the terrorists sword.
>
> Do not fall into the trap so many companies have fell into before;
> believing MS goal is peaceful co-existence. It never has been MS' goal
> and it certainly is not their goal now. MS has always proved that they
> cannot be trusted to do anything but kill the competition and promote
> more MS dependency and lock in. MONO is a good idea, but it founded on
> the premise of allowing MS to continue to exert total control. Java
> and Open Source like Python, Ruby, Perl, etc. are the only acceptable
> frameworks from which we can build an infrastructure that is totally
> independent from commercial control.
standard.
That said, see Miguel's posts about Sun and the open sourcing of Java.
Sean Kelley
been king for a long time now and will continue to reign for some time
to come. Microsoft or rather elements within Microsoft are realizing
the benefit of cross platform work. Adobe has only opened up as
little as possible in reaction to Microsoft's latest more open
strategy. The irony is not lost on the reader.
Miguel de Icaza
> NET and MONO are good examples. I personally think embracing NET was
> and is a very bad thing for Linux. The more we support MS' Java, the
> more we empower MS to control. MONO cannot and never will be where MS'
> NET is. Why? MS totally controls NET, its future, direction, features,
> etc.
Microsoft controls their product, they can not control where Mono decides to go.
Lets look at other examples, even if AT&T invented C, they had as much
"control" over C and Unix as Microsoft here. This does not mean that
third parties did not reimplement, embraced, explored, and expanded on
the original invention.
GNU C has many extensions to the language (the Linux kernel can not be
built without those). ISO C has been influenced by many third party
implementations and many improvements that have done externally.
The same applies to Unix: barely anyone runs AT&T "defined" Unix. The
early BSD fork lead the innovation charge in networking, the socket
API, file system performance and much more.
Mono has its own share of innovation, features that are Mono-specific,
you are clearly not familiar with what we have done in the Mono land,
I suggest that you get involved if you want to learn more.
But point in case: Silverlight was a technology created for the web,
we have already taken it out of the web, and used to build desktop
applications and write our own "desklet" engine, and used it as a
powerful canvas for our own applications. The list goes on.
> They have recently released their source for viewing no doubt
> hoping they can use this to promote a patent issue case against MONO,
> should the Open Source world start actually embracing MONO and begin
> to actually start building a software stack with it. Silverlight is
> giving part of this ability.
You are showing signs of paranoia described here:
http://brucebyfield.wordpress.com/2008/01/02/conspiracy-theorists-and-free-software/
Releasing source code is not new. Microsoft used to release its
source code for many components like MFC. This is a move to make the
life of software developers easier, it has nothing to do with Mono.
> I wont reply to your other two paragraphs, as I have already done so in other replies.
> I would rather see a Silverlight clone based not on NET but on Java or
> something else. That way regardless of what MS does, a non NET based
> Silverlight can adapt without having to depend upon a NET that will
> constantly change to further MS' idea of total computer and software
> dominance, which is not good for the future of computing.
A "clone" that clones the API, regardless of the language used will
have the same problem that you describe Mono/Moonlight have. So it
does not seem that you have made progress (well, at least it confirmed
that this discussion is about "My language is better than yours",
which we suspected from your email address).
Now, if you meant to say an implementation similar in spirit, but not
a "clone", then you would be correct. But then again, you would not
get to render any Silverlight content. JavaFX is probably such a
beast, so feel free to use that today in your web site.
Miguel
Miguel de Icaza
> vilifying Silverlight because it came from Redmond and therefore
> nothing even remotely related to it could ever be good, people would
> take moonlight, port to every platform and make it more ubiquitous
> than even Silverlight. I.e. Silverlight is browser only, but moonlight
> is not bound by this. I'd love a platform that used Moonlight+mono to
> write WPF styles apps across all platforms.
Agreed with the sentiment.
We have a few users that are doing exactly that today, I hope that
they will announce their products soon.
Miguel.
Miguel de Icaza
> > Here is a simple example: you take Moonlight and fork it to include a
> > useful method "createFromJson" in addition to the "createFromXaml"
> > that it has. Distribute it. Repeat for any cute innovation that you
> > want until you are tired.
>
> Another difference here: you see this as controlling, I see it as
> merely influencing. In other words, I don't think there is enough
> "bite" in the above approach to "control the destiny of Silverlight".
> But, whatever, we're both speculating, let's just see what happens
> over time.
I agree with you. The value of such a fork depends on whether the
features in such improved fork would have an impact to drive adoption
for it.
And I believe here we come full circle, if people do not like Flash or
Silverlight they are going to have to put some serious work on coming
up with something that can compete in this space.
In particular, this space at this point has a few requirements for a
new player to make a dent. This is based in large part on the status
quo of this market, so a contender must:
* Be cross platform and cross-browser.
* Improve upon the Adobe and Microsoft development tools.
* Reduce video size (bandwidth costs)
* Improve video quality.
* Bring something incredibly innovative not found on the other two.
* Provide better tooling for video pipelines than those available from
the previous two.
* Simplify the deployment of these apps.
Am not sure that those proposing the alternatives have spend much time
thinking about how to address the above issues.
Silverlight has already improved upon various of the previous areas
compared to Flash. In my opinion, without doing this, Silverlight
would not stand a chance, but as things are today Silverlight is
becoming a very strong offering.
> Consider how you brushed off my point about the Opera/Microsoft case.
> You said that case was a joke, then asked why Opera can't get market
> share comparable to Firefox, and then said, "So am not sure that
> debating things on the light of that has any merits." As if the debate
> lost merit because Opera sucks? The point I was making only
> tangentially involved Opera, how could the whole thing be shot?
The request from Opera to the EU to intervene was a joke *to begin
with* and that is what makes a debate based on this premise pointless.
Opera is a fantastic product, the problem with the complain is that
they are in a well served and saturated market. Their
differentiators are not important enough for people to bother
switching, not in the proportions that people have switched to
Firefox. Firefox proved that switching was possible, regardless of
bundling.
That is not an ad-hominem attack, I see no merit on debating Moonlight
in the context of those actions. And it is not my job to educate you
on the pros/cons of the Opera EU letter, there are some interesting
threads in OSNews, Slashdot and you can probably still find the
commentary on Techmeme if you actually want to research the topic.
Miguel.
Miguel de Icaza
> Not at all. I've never said or thought such a thing. It seems to me
> its adoption might be slower or more limited without Moonlight, based
> on comments by people who say "ah, but is it cross-platform including
> Linux?" and get the answer "Moonlight!".
So this is the core of the issue.
How slower would it be? I bet it was not a consideration for the
viewers of "Jackass 2.5".
> Jeremy Allison could have told it to you. I wish somewhere there was
> an archive of everything Microsoft's done since the early 80s, it
> would be worth your study.
There are a handful on the net, but they are typically written by
angry people, and the contents are packed with errors, so it is hard
to trust many of them.
> Of course, but that's irrelevant. I'm talking about Microsoft's
> behaviour patterns here.
You are not describing a Microsoft behavior. You are describing the
behavior of dominant players and new players in a market. The
behavior can be observed in every industry.
> We are delivering something that millions of developers want, it
> seems. Maybe there are other approaches you could have tried to
> satisfy other developers. For example, what about pitching in to help
> make Mono/C# a first-class Web scripting language?
We tried at some point with Mozilla, ask Brendan or Mike Shaver about it.
Today there are people shipping Mono in web browsers, in fact, they
deliver 60,000 downloads per day of Mono as a plugin to run code on
the browser. But still, the market dynamics prevent it from being a
well known product. I bet you have never heard of them, they have a
technology that makes every RIA platform pale today (caveat: they use
Vorbis, and that is a problem from a file size standpoint, but
ignoring that...)
As I pointed out in another post in this thread, there is much more
than just shipping the technology. A good vehicle helps, but there
are a number of factors that are required, such approach should:
* Be cross platform and cross-browser.
* Improve upon the Adobe and Microsoft development tools.
* Reduce video size (bandwidth costs)
* Improve video quality.
* Bring something incredibly innovative not found on the other two.
* Provide better tooling for video pipelines than those available from
the previous two.
* Simplify the deployment of these apps.
> Yeah, but those are clearly not part of the standard, so we can tell
> developers that we don't implement that, or they should do it
> differently in a standards-based way. Whenever Microsoft adds
> something to Silverlight that's automatically going to be part of the
> (de facto) standard.
Correct, if we are unable to implement those features, -we- Linux
users would be at a disadvantage.
And hence the imperative need to continue to work with Microsoft to
work better with the rest of us, better to work with them than work
against them.
> > > Moonlight will be in the same position as Samba and OpenOffice and
> > > Mono: always behind, always having to work harder ("on the treadmill",
> > > as Microsoft people say), always less compatible with the de facto
> > > standard. I'm amazed that you're OK with that scenario. It doesn't
> > > sound very romantic to me.
> >
> > That is because you are being defeatist.
> >
> > I used to be a sysadmin before starting Ximian, and as a Samba
> > sysadmin, Samba could do *so much more* than you could do with Windows
> > and Lan Manager that it is not even funny. There is just no
> > comparison!
>
> So what? You think the Samba complaints to the EU were just whining
> and actually they were doing just fine?
No. My point is that implementing a spec does not tie your hands to
innovation or plotting your own destiny if you choose to.
Samba has its hands tied from the same perspective that we would,
namely the market share mapping to a variation of the tree falling in
a forest and nobody hearing it.
This means that we have to innovate/improve extend in ways that keep
us compatible (or we risk wasting time).
> > As for change, I believe there are many winds of change inside
> > Microsoft. Starting with people in Scott Guthrie's group that
> > understands the value of openness. I have blogged about this
> > previously. Microsoft is not going to become the next Sun when it
> > comes to opening up code, but am happy with taking steps in the right
> > direction.
>
> I hope you're right, but as long as Steve Ballmer, Jim Allchin and
> others are in charge, I don't expect to see major changes in the way
> they compete. (Except as they bureaucratize and become generally less
> effective.)
Jim is now gone.
Miguel.
Miguel de Icaza
> Strategy 3) Implement Otherlight(TM) - i.e. implement something
> similar to Silverlight, but different.
> + Otherlight is open-source, mono team can write their own standard.
> - Take a long time and resource to design this type of software from
> scratch.
> - Since Silverlight design is good, which bits should be changed so
> that its not the same?
> - Windows developers can't easily port their code and transition to
> linux.
> + Microsoft can't play the cross platform card.
Add: "minus, Linux users are locked out of sites that use Silverlight".
In addition, bootstrapping this to be a viable competitor is a
difficult path, cut and pasting from a previous mail of mine, such
OtherLight would have to:
* Be cross platform and cross-browser.
* Improve upon the Adobe and Microsoft development tools.
* Reduce video size (bandwidth costs)
* Improve video quality.
* Bring something incredibly innovative not found on the other two.
* Provide better tooling for video pipelines than those available from
the previous two.
* Simplify the deployment of these apps.
That being said, I very much like the way you summarized this
conversation, I think those two questions are worth answering.
MIguel.
Robert O'Callahan
> On Jan 5, 2:28 pm, javama...@gmail.com wrote:
> > But lets suppose Icaza and the MONO group WERE able to keep MONO free
> > from patent infringements (which will ultimately proved increasingly
> > difficult).
>
> "But lets suppose Linus Torvalds and the kernel hackers WERE able to
> keep Linux free of patent infringements (which will ultimately proved
> increasingly difficult)."
was designing the API the kernel must implement.
This is a big problem. Even if the treadmill is spinning slowly right
now, Microsoft has many ways to make it spin as fast as they like in
the future. Ultimately they can patent something, make it a
compatibility requirement, and then tell everyone they need to license
the patent. As they threatened to do with WPF:
http://www.theregister.co.uk/2005/06/16/_microsoft_indigo/
And as they actually did do with ASF:
http://www.advogato.org/article/101.html
> The same argument can be applied to any piece of software.
specifications are written with the consensus of multiple vendors with
IP disclosure requirements and a policy that specifications will be
implementable royalty-free.
Rob
Robert O'Callahan
wrote:
> behavior of dominant players and new players in a market. The
> behavior can be observed in every industry.
> Today there are people shipping Mono in web browsers, in fact, they
> deliver 60,000 downloads per day of Mono as a plugin to run code on
> the browser. But still, the market dynamics prevent it from being a
> well known product. I bet you have never heard of them, they have a
> technology that makes every RIA platform pale today (caveat: they use
> Vorbis, and that is a problem from a file size standpoint, but
> ignoring that...)
> * Be cross platform and cross-browser.
> * Improve upon the Adobe and Microsoft development tools.
> * Reduce video size (bandwidth costs)
> * Improve video quality.
> * Bring something incredibly innovative not found on the other two.
> * Provide better tooling for video pipelines than those available from
> the previous two.
> * Simplify the deployment of these apps.
solve in free software. The rest are certainly on our agenda for the
Web platform.
> No. My point is that implementing a spec does not tie your hands to
> innovation or plotting your own destiny if you choose to.
I will now point out though, you have a problem if you add some
feature to your implementation and then Microsoft decides to add the
feature to the platform in a different way: you are forced to carry
two implementations forward, and they aren't.
> Samba has its hands tied from the same perspective that we would,
> namely the market share mapping to a variation of the tree falling in
> a forest and nobody hearing it.
platform. I want to exploit that to the maximum.
> Jim is now gone.
Good point.
I want to emphasize that most Microsoft employees I know are great
people with good intentions, so I'm not surprised you find them so.
The problem is they take orders from people who may have different
ideas. The Web standards champions at Microsoft could do nothing when
IE development was halted and developers were moved to WPF. Right now
Microsoft employees chafe under a gag order over IE8, while Web
developers scream to be told what it will do (apart from render Acid2,
yay); the gag order comes from the top, probably Steve Sinofsky. It's
those VPs you need to be converting. Although I'm not sure why they
would change, monopoly serves the shareholders well.
I think we've just about reached the end of this thread. Thanks, I
wanted to understand your thoughts and plans and now I do. I didn't
expect to change your mind, although I'll be glad if these warnings
help you later. Good luck, I hope your optimism proves right.
Rob
Robert O'Callahan
On Jan 5, 9:04 pm, greg <greg.l...@gmail.com> wrote:
> 4) Fix SVG
> - I remember Roc complaining that vendors who don't care about the
> web
> have hijacked this standard, making it very difficult to implement
> (Irony?).
anything we want.
This (or rather the Web platform of which SVG is a part) is our
strategy.
> Roc argues that users of open source platfroms should be willing to
> will make a difference to the future of the web.
Microsoft's own efforts, *then* there is nothing to lose and much to
gain from Moonlight. Until it's essential, the pain is not great.
> 1) How much pain should users of open source platforms be willing to
> take on in order to make the web a better place?
what units to measure pain in :-).
> 2) How much difference would the sacrifice likely make in terms of
> Silverlight adoption?
solution and Moonlight makes Silverlight an option for them.
> In my opinion, if enough developers like Microsoft's tools and API's
> they will make RIA's and animations for Silverlight. And html will end
> up being used just as a backup for people with mobile devices or os's
> without Silverlight. Not a nice thought - but if their tools are
> better it will likely happen.
the all-Flash route and then actually gone back to HTML/AJAX, for
interesting reasons (e.g., the difficulty of serving ad-broker ads in
Flash) that also apply to Silverlight.
> I can only see two ways to stop Silverlight adoption:
> 1) Write better or comparable vm/apis/tools, gain majority adoption,
> then standardise.
> 2) Make a large majority of developers and managers believe in the
> free software ideals, and to be willing to take a productivity hit by
> using lower quality vm/apis/tools.
exist, where free software implementations already have enough market
share to wield significant influence.
Rob
KCorax
don't run linux. Is there some relatively current live cd or virtual
machine image that can be used for demonstration purposes ?
Miguel de Icaza
> I'm looking into showcasing Moonlight for university students who
> don't run linux. Is there some relatively current live cd or virtual
> machine image that can be used for demonstration purposes ?
We have a VM that is about 2 months old, but we could upgrade Moonlight there.
Alternatively, you can contact me privately, and I can give you an
installer that contains moonlight ready to run.
Miguel.
Miguel de Icaza
> > deliver 60,000 downloads per day of Mono as a plugin to run code on
> > the browser. But still, the market dynamics prevent it from being a
> > well known product. I bet you have never heard of them, they have a
> > technology that makes every RIA platform pale today (caveat: they use
> > Vorbis, and that is a problem from a file size standpoint, but
> > ignoring that...)
>
> No I haven't, sounds interesting, URL?
Exactly what I thought and to my point.
Here is the url: http://unity3d.com
(More later, I have to run now)
Miguel
Miguel de Icaza
> Right now
> Microsoft employees chafe under a gag order over IE8, while Web
> developers scream to be told what it will do (apart from render Acid2,
> yay); the gag order comes from the top, probably Steve Sinofsky. It's
> those VPs you need to be converting. Although I'm not sure why they
> would change, monopoly serves the shareholders well.
This is entirely reasonable. They went too far in the PDC 2003 with
how much they told the world they were doing, and that has backfired
time and again when they underdeliver.
Am not an IE user, but their explanation for not talking about their
direction makes is along the above lines; I would be just as
cautious if I were them.
In addition, anyone has the right to do a "big splash" release if they
chose to. It does not map with open projects, but those are the
exception rather than the rule.
Miguel
Tzctlpc
"Licensees may also charge a licensing fee for their modified work if
they so wish".
So no pay, no code baby, no matter if you just copied it and the only
thing you did is to put a loop displaying $ signs.
Some developer could say "Oh yes, I am sharing the code all right, it
will cost you the modest amount of $1,000,000 for the CD with the
code"
MS could do so itself.
I wish the FSF would be more specifc about these issues.
On Jan 4, 10:09 am, greg <greg.l...@gmail.com> wrote:
> From Jim Hugunin's blog:
> "In addition to the Silverlight release, we've also made the full
> source code for both IronPython and all of the new DLR platform code
> available on codeplex under the BSD-style Microsoft Permissive
> License. All of that code can be downloaded today as part of the
> IronPython project at codeplex.com/ironpython."
> See:http://blogs.msdn.com/hugunin/archive/2007/04/30/a-dynamic-language-r...
>
> The MS-PL is an OSI certified license see here:http://opensource.org/licenses/ms-pl.html
>
> From Microsoft:
> "Microsoft Public License (Ms-PL) - The Ms-PL is the least restrictive
> of the Microsoft source code licenses. It allows licensees to view,
> modify, and redistribute the source code for either commercial or non-
> commercial purposes. Under the Ms-PL, licensees may change the source
> code and share it with others. Licensees may also charge a licensing
> fee for their modified work if they so wish. Microsoft uses this
> license most commonly for its developer tools, applications, and
> components."
> See:http://www.microsoft.com/resources/sharedsource/licensingbasics/share...
>
> Seem's like open source to me - or am I missing something?
>
> Cheers,
> Greg.
>
> On Jan 4, 7:32 pm, Ville Voutilainen <ville.voutilai...@gmail.com>
> wrote:
>
> > OK, it's nice that Microsoft is opening up some of its tools and
> > technologies. People are understandably wary. I would not list
> > IronPython as a thing that Microsoft open-sourced, quite the opposite.
> > The original IronPython was AFAIK available under a bsd-style license,
> > when the author of IronPython joined Microsoft, IronPython was
> > released under some sort of shared-source license. Minor step
> > backwards IMHO.
maxolas...@gmail.com
Linux, or .doc support in OO.org, or DVD playback ability in Xine.
Yes, having this support does provide some sort of legitimacy to the
patent owners of these technologies, but the fact is, without open
source implementations of these proprietary specs, using Linux would
be practically useless for most computer users.
In Linux we need to provide open source support for proprietary
standards, so that we can use these formats when necessary, yet push
open source alternatives in every way we can. Such as putting .ogg
media on our website instead of closed formates, or saving our
documents in odf format instead of .doc.
Tzctlpc
The argument about no software being safe of patents is moot of
course, but the important thing is that MS's CEO had made public
threats on this regard against Linux in particular.
Something else to consider is that any people working with this code
could not claim to have worked in "clean room" conditions, which is
essential for reverse engineering if needed. So basically anybody who
would have worked with code that could be encumbered by MS's claims
would be unable to work in the code for future maintenance of a
reverse engineered solution when the threats were made effective. How
nice is that?
If so many people are ready to trust MS's word in regards to this
"cooperation" then why they are so unwilling to remember the specific
threats made by the person dictating the direction of the company?
Miguel & Co feel of course very safe because the deal their company
made with MS, but by means of this deal and their work in direct
cooperation with MS, they could be introducing a trojan horse of the
size of Troy itself into a technology they hope will be vital.
Chaps, if you want implementation of this technology reverse
engineering is the safe answer. Anything else is exposing other people
to very specific and public threats, and although I understand
Miguel's frustration with being catching up all the time, that should
not mean we close our eyes to legitimate concerns.
If MS is such a willing partner, why don't you guys who thing this
technology is so good, ask them to make a clear statement releasing
everybody from patent litigation?
On Jan 4, 4:15 pm, jeff.stedf...@gmail.com wrote:
> On Jan 4, 8:22 am, Vexorian <vexor...@gmail.com> wrote:
>
> > Regarding your last question, these are things that prevent moonlight
> > from being extended and becoming an standard:
>
> > - Nobody needs it.
> > - Due to patent and FUD pacts, you'll need to pay MS a patent tax in
> > order to ever distribute moonlight which would make it a terrible
> > standard.
>
> Wow. Convincing argument.
>
> "Nobody needs it": Well, if that's how you want to argue... nobody /
> needs/ anything beyond food, water and air to breath. People do /want/
> it, though.
>
> Moonlight: 1 Vexorian: 0
>
> Patents: No software is safe from patents. There's nothing
> fundamentally new in Silverlight that you couldn't use prior art to
> refute. The only patents you'd have to worry about are the ones for
> the audio/video codecs, but that can be remedied in two ways: 1.
> license the codecs like everyone else (including Microsoft) has to do;
> 2. use Ogg/Theora instead.
>
> Moonlight: 2 Vexorian: 0
>
> Thanks for playing... better luck next time.
Miguel de Icaza
> could not claim to have worked in "clean room" conditions, which is
> essential for reverse engineering if needed.
Microsoft will provide us with all the specs necessary to implement
Moonlight. It is part of the collaboration agreement, the details
are somewhere at Microsoft and Novell's web sites.
In addition, the idea of "mental contamination" from such process
seems odd, specially considering that reverse engineering is allowed
for interoperability purposes. To you, I say, do not get involved in
implementing it. There are plenty of other things to do in OSS.
Miguel
Miguel de Icaza
>
> "Licensees may also charge a licensing fee for their modified work if
> they so wish".
You must be new to open source. This is a perfectly valid open source license.
> So no pay, no code baby, no matter if you just copied it and the only
> thing you did is to put a loop displaying $ signs.
You seem confused. I suggest you learn about open source licenses,
look up "BSD", "Apache" and "MIT X11" licenses on the open source web
site.
> Some developer could say "Oh yes, I am sharing the code all right, it
> will cost you the modest amount of $1,000,000 for the CD with the
> code"
Correct, and this is already the case for *all* software released
under open source/free software but not-copyleft licenses. You must
be new.
Miguel