The "delicate genius" did a short video of my deepzoom samples, I created these samples with the view to push deepzoom into the mainstream. I wanted to show people that there was more to deepzoom then a zooming picture!
(ignore the fact that I need a haircut)..
Thanks "Michael Kordahi" for taking the time to do this video!!!
Silverlight and other cool things ... 
SuperStar…. That’s the title I was looking for you. And here you are, the SuperStar!
That was an awesome video “Dude” 🙂 And it’s nice to put a face on the voice now. I want to see some of those stuff man! Great work.
..Ben
Seriously now! I’m confused with the X-ray demo. What data was being sent (pushed) to each student and at what rate and how soon the performance and usability drops as the number of students grows?
There is so much stuff in that video that you can write blog for the whole month!
Thanks Ben the samples were cool to create, the sockets sample only sends mouse co-ordinate information (at most around 5K of data per transmission) so if you do the maths if there are 1000 connected students (assuming the webserver can create that many open sockets) then if the teach moves the mouse it will send out that 5K packet of information to all 1000 socket connections. So yes the socket server needs to be a beefy server BUT that is what you need to plan for with push model applications. I’ll be setting up a demo server with the socket stuff so that people can play with it and experience it for themselves..
ps. I really believe deepzoom can drive adoption of silverlight in a big way, especially if people out there use it for things like I’ve demoed in my samples!
You’re great!
Here is my big deli ma and I’ve asked this question and have not gotten a reasonable answer. In every DZ demo that I’ve seen, I see the developer gets the pictures and uses the DZ composer and creates the new images and directory and puts it into a SL app. It’s great except with one problem. I want to write an application that uses DZ, and allows the end users to upload their images to my server and then I want a service that does what the composer does and creates the new images and folders and get the images ready for that user, all automatically (meaning me calling APIs and doing the processing). The only way that I can use DZ in large scale is to have those process be done in the back end and not the way I’ve seen so far.
From your experience with DZ, is this something we can do?
Thanks!
absolutely.. you can automate deepzoom using the sparseimagetool.exe (search for it). All my samples I created using a automation tool i created that sits ontop of that exe. The sparseimagetool.exe is what the deepzoom composer uses to create the deepzoom images.
Go forth and play with the exe Ben, you’ll grow to love it !
Wow, what a life saver. This changes everything!
Have you started writing a book yet? 🙂
Hey, don’t be stranger, you haven’t posted for ONE month man! I know it was worth it!
Really good work here, nice to see someone thinking about how to put this tech into practice in ways other than an zoomable image gallery.
Obviously the sockets solution to pushing data to clients requires that the port be open which might be an issue for alot of clients due to firewalls etc – I’ve seen a couple of mentions that “Duplex” calls to WCF services coming in SL2Beta2 which might allow a compromise (Port 80, push data). Can’t wait to get my hands on it!
Just wondering about the document example – I’m presuming you did some hand cranking to get the document pages as separate images?
Hey Bob,
Basically it depends on the Document converter, how it handles the conversion to images. I used the Adobe converter (trial version) that turns each PDF page into a jpg or png. After which I use the sparseimagetool.exe to import the collection of images as a deepzoom collection.
Basically easy sailing after the PDF conversion.
As for the sockets, your 100% correct in that the socket needs to be opened up on the clients end OR you can use port 80 on a dedicated server to serve up the socket server (which is what Im going to do for my demo server soon). Ofcourse this approach means that the server will be a dedicated socket server, no web server over port 80 for that server!
ps. Also looking forward to the WCF duplex mode! Will be heavily investigating that functionality when it comes out! 🙂
Got it, so if people were uploading a range of document types we’re in great shape for images and video and have a bit more work to do with the likes of PDF / Office Documents.
For these our current options are to use this sort of conversion for PDFs (although you then loose the benefits of vector based documents) and convert the office docs to XPS and use the XPS Viewer stuff from Delay to display – I’ve got some prototypes working using this method.
I’m wondering whether to keep bashing away at using these techniques and making them more robust or whether there is going to be any stronger built in controls for document support in SL2 or beyond. Clearly MS don’t want to bastardize their Office revenue but I can’t see a lightweight SL control for XPS viewing being anything other than a positive for them.
I’ve been spending a lot of time building an online real-time collaboration system using SL2 – was a bit gut wrenching to wake up this morning and find out that Acrobat.com had essentially turned into what I’d been trying to build! Let me know if you’d be interested in a demo, pretty rough round the edges at the moment but thinking of trying to productize by the SL2 launch proper.
Agreed on the server setup if you want to pursue sockets implementation – although it’s my understanding that the port range for SL2 is currently 4502 – 4532 which still leaves us the firewall issue, clearly this could well change as we move forwards also.
Read your encouraging post on Mesh earlier, I’ve had a couple of sleepless nights thinking about the combination between the Mesh and SL applications, particularly in this area of collaboration, although it looks like Mesh might have too much network latency to be real time enough. Even still it looks like we’re back at a similar point to when .NET was launched, where it took a lot of people a while to get a handle on what exactly it is and what it enables.
Exciting times!
Awesome stuff Jose. the integration in the website and the using deepzoom for the viewing of documents in hidable control = SUPER HOT.
on the teacher -students thing.
This is something you may be able to do once Live Mesenger is out and the Live Messenger final API is out. building a plugin for Live Messenger that works alike to the Yahoo messenger photo sharing screen. the sorting of people for a case study, course , etc could be done by the integration that Live Groups and the Windows Live wave 3 may bring later this year.
i mean. that is how i would look at it.
But i will be catching up with all this stuff in Mix essentials Mexico this week.
i meant “once Live Messenger 9 final is out (since the final Lve Messenger API should be out shortly after)”
very good …
deepzoom site
some nice work you have done there mate…
might need to commission something down the line 🙂
Pingback: Don’t worry Silverlight , you’ll always have me :) | Silverlight and other cool things …
This is one awesome blog article. Really thank you! Keep composing.
I like the helpful information you provide in your articles.
I’ll bookmark your weblog and check again here frequently. I am quite certain I’ll learn a lot of new stuff right here! Best of luck for the next!
Just wanna remark on few general things, The website style is perfect, the subject matter is really wonderful. “If a man does his best, what else is there” by George Smith Patton, Jr..
Not to hijack this thread, but, I’m looking for an investment advisor that was my parents’ retirement manager. I’m really frustrated over this at the moment because I want to work with the same agent I’m not seeing them in any of the local directories online. Carlton Financial Group – 106 Mission Ct #701 Franklin, TN 37067 phone (615) 794-2536 Does anyone know of a directory or service that makes it easy for me to search for them?