JD on EP

Flash Lite talks with J2ME

Flash Lite talks with J2ME: There were articles earlier today about Sony's Project Capuchin, but now Sony has info on their website. It seems like Sony phones will have an API so that a local Java engine can communicate with the local Flash Lite engine. They cite three use cases: a Java application triggering a Flash application; using a SWF presentation layer atop a Java processing/services layer; and intermixing SWF components within a Java application. Inter-engine communication has taken place in web browsers for awhile, but implementation differences and latency limited uses... may be different in this mobile implementation. I asked within Adobe this morning for context, and heard there will be more info after a conference next week. Sony's got some source info up now, though, to go beyond the morning's news articles.

Noteworthy injection

Noteworthy injection: User-generated content is great, but you can't trust it, and must vet it before republishing it. This weekend the Barack Obama website accepted a comment from a visitor but did not strip out angle-brackets and quotemarks. The result was a page whose new user-generated JavaScript content redirected to the Hillary Clinton website. See Wikipedia for an intro to the need of protecting your formfields from injected commands by visitors, and XSSed for additional details on the political redirect. Me, I'm hoping the next debate has a question about how each candidate feels about cross-site scripting exploits, and whether libraries like Scriptaculous should always insist upon formfield validation.... ;-)

AMP, geo-restrictions

AMP, geo-restrictions: Link goes to a machine-translation of a German PC-Welt article, about how some television shows (such as the CBS "CSI" property) can no longer be viewed in all regions via Adobe Media Player. The viewing policy is set by the content provider (such as CBS) rather than by the technology provider (Adobe), and recently CBS did enact viewing policies to respect their local agreements. I've been searching around and haven't found as many complaints as I might have expected, but it's a difficult thing to search for... if you see repercussions or have other questions then please drop a note in comments here and I'll work on it, thanks.

Label debates

Label debates: I don't usually link to Sys-Con, or to Scott Barnes, but this collection of paragraphs asserting that "'RIA' is slowly fading in terms of its definition" is a piece of almost offensively obtuse obscurantism. Rich Internet Applications combine serverside-processing (think Allaire) and clientside-processing (think Macromedia) for something richer than just text and images. Microsoft was slow to arrive on the scene, but they've understood a part of it recently... they even hired Scott with a literal "RIA" in his job title, for their as-yet-unshipped embrace/extend rerun. But now: "The team with the biggest horde will own the definition." As when Microsoft woke up a few years ago and tried to diffuse the idea as "Rich Interactive Applications", muddling things now with "Rich Client Platform" attempts to confuse their core customer base, preventing their growth outside of the Redmond stack. It won't affect the underlying dynamics in the long term, but will remain a time-wasting nuisance in the short term.

]]> (Trolls: Please sign in with a verifiable identity before calling me childish and mean for my borg-defiance, thanks.)

Creative accessibility

Creative accessibility: "Freedom of the press is guaranteed only to those who own one." [A.J.Liebling] Twenty years ago desktop publishing reduced the cost to tens of thousands of dollars. A decade later web publishing helped more people publish and distribute text and images, costing only a computer, its software, and a connection. Now, rich-media RIAs let anyone edit images, video, audio and presentations, without any investment in software at all. Third world or first, urban or rural, egghead or not, it doesn't matter anymore -- if you've got a computer and a connection, you can create and publish your story to the world. At PBS, Jennifer Woodard Maderazo surveys the range of accessible creative tools today, and compares it to the costs of the recent past. The professional toolsets are blazing even newer territory (time-based imaging is about to get real scary, eg), but the increasing accessibility of the tools of creation means that Liebling's guarantee is becoming less and less a restriction every day....

]]> Action items:

We've got to get this functionality to cheaper mobile devices, because most people will never own a computer. The Flash Lite runtime is getting there, across the more affluent sections, and the AIR runtime will eventually provide a solid baseline of functionality. Right now, though, digital creation still requires a computer, a connection, and usually some level of English skills.
We've got to figure out ways to beat Sturgeon's Law and filter out the bad talk from the good. If you're trying to learn something, how can you pull that information from the efforts of people to sell you something, or the casual rants of the popular? We've got to reduce the reading costs, make it easier to find the desired info, while still surprising with the unknown-yet-useful. Filtering will probably be a harder problem than creation.

Adobe to acquire Macromedia

Adobe to acquire Macromedia: Announcement went out three years ago today. Seems longer. Why, back then we didn't even have YouTube... things have changed a lot since then.

Bad Reporter, No AdSense

Bad Reporter, No AdSense: The link goes to a Google search on pages indexed within the past 24 hours on "'mark dowd' adobe"... currently gives 40 pages about yesterday's issue where blogs/newspapers highlighted an old Player vulnerability. Now try the same term and add "9.0.124", to see how many of these articles noted that the vulnerability is already addressed, and readers should update. Result? Zip. These commercial writers are selling fear, and not serving their readers. They are the security problem, themselves.

SEO reality

SEO reality: Neat points raised by Google's VP of search quality, Udi Manber: "I wish people would put more effort into thinking about how other people will find them and putting the right keywords onto their pages... The content provider should think about how users will look for their content, and the user should think about what words people use to write about their content... You should think about what you expect to see in the actual page and search for that." Full bodytext (much less database text) is only rarely needed to satisfy a search query... usually you anticipate which terms your audience will use when trying to find your service, and on which you can reasonably compete to get on that first page of results. If your URL, TITLE, metadata and bodytext all include these terms, and if you can get external links with these terms as inbound anchor text, then you've got a good chance of being found by your audience as they will tend to search for you. I've got prior examples for rutabagas and flowers, and hit that "visualize the page on which the info lives" tip back in 2003. SEO experts may be suspect, but I'm happy to see that the people inside Google agree that the basics are much simpler: figure out how people will try to find you, and optimize for that.

Silverlight tripling

Silverlight tripling: I hate giving Microsoft public-relations people extra exposure, but I spent enough time researching it that I figure others may want to know the meat behind the story too. The headlines go "Silverlight to triple marketshare" and such, but Katherine Noyes of E-Commerce Times has some of the details on that non-public Evans Data report: "In a poll of [400] developers who work with Web 2.0 technologies, Evans Data asked both what the developers currently use for rich Internet applications as well as what they plan to use in the next year to year and a half." From what's reported of the results, forty of them were investigating Silverlight today, and 120 of them plan to do so by 2010. No word on how many actual public-facing projects were under development. The fact that only a third seem to have responded to the prior massive publicity seems small to me, but you know my perspective. ;-) Anyway, it's a positive headline for Microsoft, even though it risks overestimating eventual realities. If I get a more direct link to the Evans Data study and wording I'll update this post.

"Flash vulnerability" story

"Flash vulnerability" story: I'm bumping this up to my weblog, because OS News requires membership for comments, and their source, Thomas Ptacek, has not yet published the comment I submitted. The Mark Dowd paper describes an issue which was addressed in the current Player, v9.0.124. None of the numerous paragraphs describing the horrors seems to mention this, and because these blogs don't support open comments, they may not hear unless they get publicly called out on it. It'd be better if they had open conversations on their weblogs, though... would serve their readers better.

]]> Update 9pm PDT: The Ptacek/Matasano link got picked up by Microsoft's Larry Osterman, via Robert Hensing. Neither has advised their readers that this vulnerability is addressed in the current Player. I do not see the full story in comments at OSNews, although Ben Lucyk got it in a trackback there. (Thanks, Ben!)

Check out the comments at Matasano and OSNews... lots of "proprietary garbage" type of prejudice. The reality is that these people are not harvesting information effectively, not analyzing their harvested information effectively, and not responding to feedback effectively.

Update Thu Apr17 8am PDT: ZDNet Security Blog ran with the story last night. The guy spent paragraphs writing about basketball games and his mother, but never even checked Adobe sources to see the problem was already addressed. He is part of the problem which must be fixed... our world is taken up too much by those who speak too much, yet do not listen, do not question.

(Thanks to "Skila", a member of OSNews, who added in comments "This was fixed in the latest version of Flash Player - released 8 April 2008 so this is olds not news.")

Followup: I got an internal email last night that the ZDNet reporter did mention "new version" down towards the bottom of his text.

Update Sat Apr 19, noon PDT: It seems that most of the conversation now is focusing on the vulnerability in coding practices, rather than the Flash aspects... Computerworld had the "Adobe already fixed it" datum as the first sentence in the fourth paragraph, and with this highlighting, subsequent reporters have followed suit. Even the Slashdot discussion is more about the coding than about the Player.

I want to emphasize that the original discoverer, Mark Dowd, did act in good faith -- he notified Adobe security, and published his whitepaper only after the Player changes were public. He helped everyone by handling this the way he did. (I also understand how the early bloggers were excited by the coding acrobatics, but I wish they had clearly advised concerned readers to keep their software current. The increasing moderation of useful blog comments is a separate issue. No blame, just room for increased openness.)

Update Mon Apr 21 8am PDT: Most of the followup reports do a little bit of research, but today's BBC account is another lengthy personal reaction to the Matasano paper and the new type of coding exploit, and despite its length and extraneous details, does not advise readers that they should just update to the software already available. (Meanwhile, in comments below, the original popularizer wants me to retract that I submitted a comment there which was not published, even though he hasn't published it yet, nor amended the article to include the vital non-inflammatory news that the vulnerability was addressed before the publicity.)

AIR testimonial

AIR testimonial: I know lots of people like it, but I like how it solved a problem here. Galen Gruman of InfoWorld had some Ajax applications, and needed to make a desktop widget, and: "The weekend before I was due to deliver my working prototypes of the Windows Sentinel Web and mobile front ends, I saw that the Adobe AIR SDK and the companion plug-ins for Dreamweaver and Flash had just been released for download. I figured, 'Let's see what AIR can do.' Less than two hours later, I was able to essentially export my mobile Web pages into desktop widgets. Better, Adobe AIR produces a single executable that runs on Windows XP, Windows Vista, and Mac OS X Leopard -- and, soon, Linux." Even better is his conclusion: "I think most organizations will discover that Dreamweaver and Flash have magically become an easy lightweight development platform that a wider range of employees can put to good effect than could ever use a full-blown IDE." Lots more people can do lots more stuff. That's good. :)

Realworld HD H.264 support

Realworld HD H.264 support: In March, 4500 consumers were tested for viewing different types of media files. 2780 of them had already installed Adobe Flash Player 9.0.115. That's 62% of today's computers, supporting no-hassle high definition playback of H.264 video. Considering this browser plugin was released in December, and the audit was conducted in March, then it's an easy choice for realworld use today. (Same goes for the persistent framework caching in Flex 3, too.) [via Justin Everett-Church]

]]>
Update: From the first wave of comments, it looks like there's some of the classic confusion over what sampling is, how it works, how to test it, and how to counter it. You can search back, quarter by quarter for years, to hear similar objections addressed.

And, of course, objections are stronger when they are signed with a real name and address, and when "you can't be right because our site is less" uses at least a named site and methodology. Right now, those anonymous objections seem more threatened than informative.

The Millward-Brown and NPD consumer audits have a history of accuracy of overall consumer trends. They do not represent each site's individual audience, but the overall consumer technographics.They are checked against server traffic as well. If you believe that this quarter is somehow less accurate than previous quarters, previous years, then it might be useful to suggest what suddenly changed.

.NET devs as your competitors?

.NET devs as your competitors? I would have left this as a comment at Don Burnett's blog, but it requires Google membership. Don riffs off of a March appraisal of Microsoft's beta Silverlight browser plugin from Jens Brynildsen, and touches a number of points, but the one that stood out for me was "The .Net Programming Community is now your competition." I'm not sure... I read their forums and blogposts, and see distinct differences between the .Net community and the Flash-oriented communities. I don't think the two groups of people would start similar types of projects. Most of the .NET developers I know of are in constrained environments, and don't deliver to the open web. Flash, Flex and AIR people grew up with delivering to diverse audiences. I see differences between the two groups of people. You might compete with a .NET staff if you develop for an intranet, but do .NET-style workgroups have what it takes to excite the audiences on the open web? Open question, from what I've seen so far. You have any other thoughts on this...?

Central-ish AIR

Central-ish AIR: Fred Wilson, of "A VC" blog, likes AIR, but is seeking to bring together multiple services into a single window, sort of like the old Macromedia Central did: "The problem with AIR is that unlike Firefox, which allows for tabbed browsing, there is no way to launch one AIR client and get multiple services. I'd like to have Twhirl, AlertThingy, and Shifd (another cool AIR client) all running at the same time in different tabs." Comments point out that AIR is a way for developers to create standalone applications, rather than a way for consumers to remix application interfaces... a good read. But then I started wondering about AIR apps which combined different webpage interfaces already in existence, so that you make a "multi-tabbed" application from existing applications. Would that end up being like Pirated-Sites.com, though? We haven't really discussed reuse of interfaces created elsewhere by others. I'm not sure of this whole area... does this bring any thoughts to mind for you...?

Dead blog comments

Dead blog comments: Weblogs are pretty broken. Some writers read only through RSS and never pick up comments. But lots of blog owners get lazy on spam handling, and default-off publishing their replies. I've got a bunch of stuff I've taken the time to type, but which never got clicked-for-approval by the blog owner. No real info here -- this at-twice-remove type of conversation doesn't work well -- but if the people requesting such comment won't acknowledge listeners after they speak, then I'll just post comments here instead.

]]>

041308
http://www.embeddedflash.com/?p=60
Update: Now published.

Hi, I'm not comfortable with those WMP/QT comparisons either:
http://weblogs.macromedia.com/jd/archives/2008/04/rashomon_amp.cfm

AMP (Adobe Media Player) is an application atop AIR (the Adobe Integrated Runtime), which includes the WebKit HTML/JS engine and the Adobe Flash Player. That's why AMP is a way to manage subscriptions for video encoded the three codecs Player includes: Sorenson, On2, and H.264. There are definitely other ways to encode video too!

Mobile AIR is definitely a goal, but is currently MacOS, Windows, and Linux (alpha).

No extension model, but I'm not sure what the actual desire is.

jd/adobe

041308
http://newteevee.com/2008/04/13/secret-of-the-iplayers-success-no-drm/

"Adobe’s latest Flash version 9 is supporting encrypted media streams in order to lock out third party players and stream capturing applications."

It's more like creators can now create encrypted streams if they want... the goal is to offer a variety of contracts between consumer and creator, so that big-budget video can get a return. Adobe Flash Player 9 also renders H.264 video too. There's a range of choices.

"iPlayer users however have reported that the BBC is using Flash version 8 for their web streams since December. The broadcaster also started to reencode and optimize all of its video for Flash 7 in order to make the Wii version work since the Wii’s Flash Lite player doesn’t support Flash 9 yet."

Lengthy info on BBC production workflows here... it's amazing!
http://www.bbc.co.uk/blogs/bbcinternet/2008/03/bbc_iplayer_on_iphone_behind_t.html

(Adobe Flash Player 9 includes three different video decoders: Sorenson, On2, and most recently H.264. The mobile engine, Flash Lite, is smaller and passes off video rendering to the device's own hardware-based video decoder. We're getting there.... ;-)

Re "Secrets of Success": I'm not sure that current "marketshare" would be a measure of success, so much as long-term sustainability of continuing video production would be. We need a variety of types of contracts between video's creators and consumers, so that each can reach their best choices.

jd/adobe

041208
http://www.chuckstar.com/blog/?p=160

"Why, Silverlight is so much like Flash, it even keeps 'Adobe Flash' in its own context menu..." ;-) j/k

MLB.com still uses Window Media Server on the backend, and still uses tons of Flash for menus, data-driven displays, Gameday, more. This year they added support for the Microsoft Silverlight browser plug-in, in addition to their existing support for Windows Media Player for local viewing.

When they've got a video asset stuck in Windows-Media formats (instead of H264 or such), then they surround the Silverlight video client with their normal Flash displays. "Silverlight" is added to the SWF's context-menu instructions in the normal way.

More info:
http://www.infoq.com/interviews/mlb-silverlight
http://weblogs.macromedia.com/jd/archives/2008/03/mlbcom_2008.cfm

For the plugin versioning, if you ever install a Microsoft Silverlight plugin, it defaults to silent auto-install of updates:
http://www.microsoft.com/silverlight/resources/privacy.aspx

jd/adobe

041208
http://neilmiddleton.com/2008/04/12/scott-barnes-talks-coldfusion/

[I didn't copy the text before hitting "Submit", but it was basically "What's your own actual concern, in whatever Scott Barnes may have happened to type?"]

040908
http://www.techcrunch.com/2008/04/08/adobe-launches-media-player-adobe-tv/

Geographic restrictions would be set by policy files by the video owners. I'm not sure if there's general documentation yet on these abilities, and don't know specifics about how CBS sets their permissions, sorry.

jd/adobe