Comments

• Jim Powers

  2007-12-03 18:19:58

  Ya know, two can play this little game. A number of years ago I exchanged some emails with a FLOSS "big-wig" regarding, the at the time fiasco in the Linux kernel development community because Linux was using BitKeeper. The owner of BitKeeper threatened to change the protocols deliberately if work by FLOSS developers continued to try to reverse-engineer the current protocols. My comment was that that was a clear indication for the community to get out of the use of BitKeeper and get on its own two feet, so to speak. Eventually that did happen, we fortunately got git out of it. But in the context of my e-mail I suggested that FLOSS needs to take a more active role LEADING in the development of tools, protocols, etc. For instance, I suggested all open source browsers should create extensions for HTML only found in open source browsers. Clearly there would be no patent implications so MS ~could~ chase after it if they wanted to, and getting the W3C to accept the extensions into their standards should prove relatively easily done since the implementation would be under a FLOSS license and there would be no patent encumbrances by default.
  
  Perhaps it is time for the same mentality for a "rich" player. Design one that is really good, works on all FLOSS browsers and let MS play catch-up. I don't think that it would be too hard. We already have a bunch of sorta-decent players (they're good players, but their browser integration is ~eh~). Add some scripting tools, debugging, etc. Probably doable with ECMAscript or some such. The point is make it a high-priority FLOSS project and it will get done, and done well, why chase after Adobe (gnash) and Microsoft (Silverlight), when it is perfectly reasonable to think that FLOSS can do it better. Make them chase us, not the other way around.

• Roy Schestowitz

  2007-12-03 18:28:11

  I suggested all open source browsers should create extensions for HTML only found in open source browsers. Clearly there would be no patent implications so MS ~could~ chase after it if they wanted to, and getting the W3C to accept the extensions into their standards should prove relatively easily done since the implementation would be under a FLOSS license and there would be no patent encumbrances by default.

  
  
  To an extent, you could argue that Mozilla and Google do this already. In fact, one strand of work supports off-line Web-based applications support, which helps GNU/Linux a great deal (portability).
  
  

  The point is make it a high-priority FLOSS project and it will get done, and done well, why chase after Adobe (gnash) and Microsoft (Silverlight), when it is perfectly reasonable to think that FLOSS can do it better. Make them chase us, not the other way around.

  
  
  Sounds like the ODF of the World Wide Web. Sounds reasonable. There is already a project called Curl, IIRC, which is truly open source. It cannot receive as much exposure (and thus acceptance, embrace by developers) as projects from the 'big boys' though.

• John Drinkwater

  2007-12-03 18:38:34

  Jim: See <video> & <audio> tags for Ogg support, SVG use, Canvas 3D using OpenGL ES, there’s more than enough OS/standards pull in web browsers, we just have to make sure Apple and Microsoft don’t hold it back.

• Roy Schestowitz

  2007-12-03 18:47:53

  John,
  
  Yes, I can recall some HTML 5.0 extensions, Opera support, and Ogg Vorbis/Theora. I think I wrote about this before. The real challenge is getting developers to make such stuff widespread. I still try very hard (outside this Web site) to get the message across that Adobe Flash is proprietary and Silverlight is Windows-centric (Vista already offers XAML 'enhancements').

• Jim Powers

  2007-12-03 19:51:30

  Well, a FLOSS Flash replacement is one of the FSF high-priority projects.
  
  Regarding curl, my recollection is that it only recently became open source (as far ad the players and such). More importantly it seems like the infrastructure for "things like flash" should just get built into the browser. Supposedly SVG is supposed to do all that but for some reason (outside of KDE, Gnome ~somewhat~, and token FireFox work) it is needlessly languishing. Make it first-class and rocking! Make people /want/ it.
  
  As far as Ogg I love Ogg (and theora), all my music is in OGG, Theora is another story, again a bit languishing. I like the fact that some recent Eben Molgen videos were released in Theora, but "all that FLOSS A/V smack-fu" needs to take more of a center stage. For instance, Mozilla should bundle Theora and Ogg playback plugins by default. Clearly under Linux and such it is merely assumed that andy distro worth installing will give you this for free.
  
  Making a YouTube video playback support with an all FLOSS stack should be easy for EVERYONE, not just hacks like me that are willing to to through the pain of encoding in Therora (Thoggen don't cut it folks, but it's close).
  
  It should be something like this: See a flash "thingy" be able to build it on the all FLOSS stack in a weekend. Clearly this needs to diffuse into the tools that people already know how to use, but dammit! We need an "Apache project" of rich web development, a high-profile project that sits on top of an enviable stack, FLOSS top-to-bottom.

• Hmmm

  2007-12-07 17:48:02

  The problem is that FLOSS lacks the backing to do this... It's going to be hard enough for Microsoft with it's billions of dollars to replace Flash with Silverlight... how is FLOSS going to replace Flash/Silverlight with something that it creates?
  
  Maybe I'm wrong in thinking that it's hopeless, but it sure feels that way.
  
  That said, I'd still support the effort in any way I could... but first things first, someone actually needs to DO it.
  
  The problem with FLOSS community is they never seem to get off their butts to do anything but follow the leaders, and it's quite sad.

• Hmmm

  2007-12-07 17:50:18

  Oh, and keep in mind that and tags are not enough to replace Flash/Silverlight which do a whole lot more than simple media, though it would be a start and far simpler for web developers to use if all they wanted to do was play a video in a web browser.
  
  Also, as far as I understand it, SVG is not as powerful as either Silverlight or Flash, so while it may be a start - it would need to be extended/improved.

• Hmmm

  2007-12-07 17:51:41

  "Oh, and keep in mind that and tags" was meant to say:
  
  "Oh, and keep in mind that <audio> and <video> tags"
  
  but I had forgotten to encode the angle brackets

• Roy Schestowitz

  2007-12-07 18:11:21

  Features are introduced gradually and standards evolve with the consent of various parties that have them implemented.
  
  A 'big bang' approach to producing something and then shipping a binary blob is easy, That's not the way to go though. Consider the way Microsoft Office evolved over time and the great difficulty of formulating the work as a spec.
  
  Some people have criticised ODF for "not having as much functionality" (as OOXML). The same goes for HTML+CSS+JS compared to Flash. It is a similar scenario where one has to step back and build upon shared work in an elegant fashion.

• Jim Powers

  2007-12-07 18:34:19

  There are a number of projects already out there that touch much of what would be necessary to put Solver/Moon-light and perhaps even flash out of business.
  
  Having already said that it seems like this "stuff" should already be in the browser here's what FLOSS browser developers could do:
  
  1. Leverage HTML/CSS/JavaScript/SVG. Add built-in (i.e. native, at native speeds) multimedia functionality that is exposed via a well-defined, not yet "standard" JS api. The standardization will some in time, but the extensions NOW are what's valuable. I could easily see Mozilla, KDE, and Gnome getting together on this via an organization like freedesktop.org. Start out small and potentially flawed, but get started quickly. each of the FLOSS browser implementing a spec published through the freedesktop.org umbrella, like DBUS and such.
  
  2. Get the major FLOSS media players involved from day one so XINE, MPlayer, and VLC are in the loop day one. come up with an api for managing payers and a rendering model that allows the players to be used in clever ways in a powerful rendering model.
  
  3. Get a useful rendering API in place for JavaScript. Clearly the SVG renderer can be built on top of this. But something like native Cairo API with some asset management thrown in on top.
  
  4. Come up with a "blob" format for all of this. Perhaps an extension of ODF? or another XML-based blob format? Maybe even simpler, zip would also be fine.
  
  5. Finally, make sure that this "stack" is "instant-on" in all FLOSS browsers, and that the stack FLOSS top-to-bottom.
  
  Now, looking at the landscape of what's out there already it REALLY doesn't seem like we're all that from making it a reality. If the major players could be enticed to start this soon we could see the first fruits of this work in 2008. By sometime in 2009 Silerverlight and Flash could become irrelevant.
  
  It is the right thing to do.

• Roy Schestowitz

  2007-12-07 18:54:55

  4. Come up with a “blob” format for all of this. Perhaps an extension of ODF? or another XML-based blob format? Maybe even simpler, zip would also be fine.

  
  
  I suppose that some might argue that WPF and XAML already do something similar, control of 'extensions' aside. I've had these argument before. Another major issue is the codecs.
  
  

  By sometime in 2009 Silerverlight and Flash could become irrelevant.

  
  
  I'm sure that the residue (old sites, old skills, old products) will ensure it lasts for much longer than that. I built a site using Flash when I was 19 and it's still pretty much unchanged (heaps of Flash)

• Jim Powers

  2007-12-07 19:05:56

  Well, XAML is irrelevant as far as I'm concerned. As far as other blob formats I have no real opinion. Jar files are essentially ZIPs so certainly that could work. Codecs? FLOSS codes by default. That is the point of having a FLOSS top-to-bottom thing like this right?

• Roy Schestowitz

  2007-12-07 19:32:18

  Yes, I agree. To be honest with you, I also feel a little guilty talking about this. Despite having a Software Engineering degree, I haven't been coding much recently and I crave going back to doing projects rather than writing about their need, especially when we're talking Free software. This usually means volunteer work, but as Sun Microsystems revealed in India a couple of days ago (not for the first time), it's willing to sponsor such stuff. JavaFX comes to mind, but it isn't exactly the same thing.

• Hmmm

  2007-12-07 22:08:19

  Another thing to keep in mind is that in order for a Silverlight/Flash-killer to become popular among the broader WWW (and not just a niche thing limited to the FLOSS community) is for a really nice GUI tool to create such sites to be written and made cross-platform. These web developers need to be able to simply drag & drop things onto a page, say "slide from here...", drag-to-new-location, "...to here", click Save, and be done.
  
  My point is that creating alternatives to Silverlight/Flash using a FLOSS browser plugin (or extensions) is not going to be enough to win the war. You have to make the average web developer /want/ to use it. And you have to make it easy for them to create sites using it.
  
  Otherwise you just end up preaching to the choir...
  
  Actually... I just thought of something: I think the solution needs to be a plugin that will work with even non-FLOSS browsers so that web developers don't have the excuse of "oh, it's nice and all... but doesn't work for users using IE7" or whatever.

• Roy Schestowitz

  2007-12-07 22:19:04

  That is another issue. IE hasn't the best track record when it comes to properly supporting cutting-edge features that facilitate Web service -- you know, those sorts of 'GUI thingies' that Bill Gates warned about in the mid-nineties when he decided to 'tilt Netscape in the death spiral'. Regarding [Silver|Moon]light, we need to educate people about this issue. Yesterday:
  
  

  As part of Novell’s arrangement with Microsoft, Microsoft also is providing financial incentives to Novell to get Silverlight on Linux, de Icaza added.
  
  “Microsoft agreed to pay all the licensing and patent fees for redistributing the (Silverlight) codecs,” de Icaza said. “We don’t have the Microsoft codecs for Silverlight now. So we can not yet do streaming. … But it’s coming.”

  
  
  Yay. "Patents", "fees", and Microsoft's "financial incentives" for Novell. A recipe for another digital dark age on the Web (GNU/Linux users denied access, as was often the case before Mozilla Firefox came along and gained attraction). I hope you don't mind me being cynical here. ;-)

• Hmmm

  2007-12-07 22:40:01

  Well, to be fair, the patents/fees are for MPEG-LA - we've been living in that "dark age" for over a decade now, it's nothing new.
  
  And with Microsoft paying the bill to give all Linux users using Moonlight the right to view media content, seems like a win to me.
  
  Personally, I find it to be a brilliant hack to get Microsoft to pay for: 1. the MPEG-LA license fees for Moonlight, allowing any/all Linux users to legally be able to view MPEG-LA content (granted, only via Moonlight - but still) 2. the development of Moonlight (Microsoft paying for the development of Linux software!!)
  
  Granted, it would be nicer if the media format was OGG, but if I for some reason discover a site I want to view which requires Silverlight and install Moonlight, I'm sure glad I don't have to worry about getting sued by MPEG-LA.
  
  Sadly I couldn't say that when I listened in to the Linux.com blogtalkradio event when I was forced to use unlicensed patent-encumbered software (e.g. mplayer).

• Roy Schestowitz

  2007-12-07 23:10:52

  An Internet troll who goes by the name of "eet" begs to say something on this issue, but since he's banned, I sadly cannot display the comment here.
  
  I'm beginning to suspect that this person is nothing more than a Microsoft/Novell marketer/evangelist (i.e. shill) with Web proxies around the world. I don't know if you've followed this site long enough to know about this. Anyway, your input is appreciated as its technical and does not contain personal attacks and insults.
  
  "eet", since I know you're reading this, you can kindly stop bothering to comment. Your input lost its value after about 100 personal attacks.

• Jim Powers

  2007-12-07 23:39:03

  Agreed about "eet" sadly. It seems clear to be a professional instigator.

• Roy Schestowitz

  2007-12-08 03:05:15

  Jim, I've arbitrarily just linked to a blog post that I spotted about 3 days ago and while testing the URL I suddenly found a little friend of ours in the comments section...
  
  http://iandefor.wordpress.com/2007/12/05/the-inevitable-annoying-quirks-opensuse-103/
  
  Look who's there. He also attacks Radu whenever he mentions SUSE or Novell. He got banned. That appears to be the only scenario where he gets active/proactive on the Web. He's protective of Novell and he has proxies. I can't help but think about cases where Novell does its 'thing' (that's just among other posts on this subject).
  
  Then, come to consider all those retaliation messages about 'paranoia', 'tinfoil hats', and 'black helicopters' (shoot the messenger tactics). But of course, such ideas are just creations of our on imagination. Or, is it truly so?

• Jim Powers

  2007-12-08 05:22:39

  Ok, it may be more complicated than hired gun (shill). The behaviors so far presented seem also along the lines of maniacal fanboi. I've read about other cases where posters have gone as far as to threaten people's lives because of criticism about a particular distribution.
  
  I probably put eet in the "pretty damn smart" category, he knows something about Linux, his actions can have two causes:
  
  1. He's a "hired gun" paid to patrol the net looking for disparaging remarks about Novell/SuSE/OpenSuSE and "do something". In the case of the link above the "do something" was to educate. In the case of this site his "do something" is to cause disruption and discredit the site.
  
  2. He's an independent agent with a strong vested interest in SuSE/OpenSuSE. In this case he/she/it may not be responding to criticism of Novell appropriately. Since he feels that he is being shutout he begins to act more and more irrationally. The fact that he keeps trying, like an obsessive, is a bit worrying in this context: he may be emotionally or mentally unstable which could escalate into something worse.
  
  This site's job is to provide a meaningful analysis of the cirque de Novell. Much of that analysis is based on the data reported on the Web and elsewhere, but painting a particular picture does require speculating about the motives of Novell and its partners. Of course this speculation can be wrong, nobody is denying that, but so far this is not fake moon landing/X-Files conspiracy stuff. The actions of Novell and its partners are a matter of public record. Much of the analysis presented here (and elsewhere) is directly derived from "particular statements made on the record" like the Halloween documents, for instance. It's OK to be skeptical of the skeptic, there are ways to address this such as making factual corrections, offering alternative analysis (with backup), etc. In this regard eet is not helping himself (or others). Eet can redeem himself, even as critic, by addressing factual errors, or offering alternative analysis. Much of his commentary has degenerated into snide drive-by comments motivated by anger with no rational core, and the rational core is what is needed.
  
  For instance: this site is accused of engaging in "wild speculation". There is a hint of truth in this because "we" are trying to deduce the "plan" as understood by the "insiders" in this particular issue. Such a plan is not going to be made generally public, and unless someone "on the inside" of all of this can bring forth a "smoking gun" piece of evidence we are left with nothing but Novell et als observable behavior and statements to derive the nature of "the plan". Now, eet seems to have come to the conclusion that BN is making mountains out of molehills. This would be a reasonable conclusion if it were not for the fact that Microsoft is so clearly involved in all of this. With a public offensive on all things FLOSS (certainly FLOSS as we know it), Microsoft's motives have to be held suspect. To think otherwise is historically very foolish. Since Novell is now partnering with FLOSS' deadliest opponent Novell's actions cannot be assumed to be FLOSS friendly without evidence. Clearly Novell could act in ways that would be atypical for Microsoft w.r.t. FLOSS, but that does not seem to be the case. Given all this what is a reasonable way to proceed?
  
  a) Admit that we've all had crazy thoughts and really there's "nothing there"
  
  b) Continue to be vigilant and continue analyze Novell and its partners over time to understand the "true" nature of this partnership.
  
  I would say that (b) is both a rational and justified position to take. Now eet would of course say "surprise, surprise, friendly poster agrees with objective of blog". Well, it's not exactly a tautology: there is no guarantee that I have agreed with all the previous posts or will agree to all future posts. I have my own sources of information independent of this site and decide for myself whether I agree or not with what is published here. eet, and his ilk, would do everyone a service if they would formulate their posts into something that can be discussed, rather then ad hominum attacks or drive-by nasty comments meant to irritate but not illuminate.

• Roy Schestowitz

  2007-12-08 06:29:52

  Excellent assessment.
  
  Well, I can only imagine that the pitch of this Web site will be toned down over time in order for the site to seem more welcoming and credible (where information always overrides emotion).
  
  

  eet, and his ilk, would do everyone a service if they would formulate their posts into something that can be discussed, rather then ad hominum attacks or drive-by nasty comments meant to irritate but not illuminate.

  
  
  It has to be stopped when readers (commenters) got attacked. I can take the insults and even the slander because I am used to it, but some people that I know fear participating in forums where they see others getting attacked. They will never show a sign of presence in hostile territories, so to speak.
  
  If a site is protective of non-trolls, then it's probably bound to encourage more discussion. Your advice along the way has helped a lot in that regard and while eet keeps trying to comment every week, he only speaks to one person.