18.5.06

XTech - day 2 - part 3

Developing for the Personal InfoCloud

This presentation is actually called "Developing for The Come to Me Web"...

  • (past) I Go Get Web

    • We sought "their" information
    • Focus on content provided
    • One device
    • One use
    • Proprietary formats
    • Findability focus
    • Development metaphor was navigation
  • (now) Come To Me Web

    • "My information" is found or created
    • Focus on person using
    • Reuse
    • Attracting and keeping attracted (we want info about some stuff, but not about some other stuff)
    • Across devices
    • Open formats
    • Refindability focus
    • Development metaphor is attraction
  • Personal InfoCloud
    • This is the information one person want to have allways with him
    • As a matter of fact there isn't one person InfoCloud but four:
      • Personal InfoCloud
      • Local InfoCloud
      • Global InfoCloud
      • External InfoCloud
    • We have to make an effort to move valuable data from the external infocloud to one of the others
    • Developing to the Personal InfoCloud
      • First of all, we have to think allways about the person, he's what matters
      • Access to information
      • Organization of the information
      • Aware
        • Task (what's the data meant to be used)
        • Action (how can you use it)
        • Context (the number of disambiguities on Wikipedia shows us the importance of this)
    • Model of Attraction
      • How to attract the information to yourself?
        • The Intellectual Receptor
        • The Perceptual Receptor
        • Mechanical Receptor
        • Physical Receptor
    • Personal Information Workflow
      • Privacy...
      • People get info from local and global infoclouds and move data to the personal infocloud by:
        • Seeking
        • Recognizing
        • Retaining/Storing
        • Using/Creating
        • Following
    • Info Reuse
      • Mobility for data is really important... Thig guy should read about Ambient Networks
    • Information is needed, everywhere, with context
    • Build around people
    • Understanding Content: Assess what is provided (caution with this, have allways in mind that information is reused)
    • Assemble information for use:
      • Learn what people do with the information
      • Various use contexts
      • Usable formats for content object type
    • Rethhinking our information
    • External Structures
    • External Storage (how do we deal with data updates?)
    • Develop with Standasrds
      • Information offerings:
        • Structured
        • Flexible
        • Open Standards
        • Proprirtary or Applicvation-based Standards
        • API Friendly
      • Offer more than one option
    • InfoCloud sollutions (his contact)


Lunch time

So, I had lunch with this really cool people from Opera, who got me preety excited to attend to the presentations I had already decided to see on Opera stuff on mobile devices. Now, and before the first presentation of the afternoon, I'll try to check my e-mail...

The Intelligent Design of Microformats

So, this is the first presentation in the Foyer room I'm attending. The speaker is Ryan King, from Technorati.
The presentation can be downloaded here.


  • How to design and create formats.

  • theryanking.com

  • Technology

    • Conservative
    • add to the existing Web
    • minimum invention

  • Social

    • Promote, encourage, buid...

      • convergence
      • openness
    • not a...
      • standards body
      • proprietary project

  • Elevator Pitch - Designed for humans first
  • data format design - this isn't software development, you can't "teach" how to analyticly design a data format
  • User Centered Design
    • "web of data"
    • we need people to build the web of data
  • designing microformats
    • experiment - easy and cool, because people are open, and share thoughts, snippets,...
    • stread
    • iterate
    • watch
  • semantic markup - maintain is preery easy; converge to best practices too
  • Problem statement
    • identify the problem
    • are there any simpler problems? Can you simplify your original problem by dividing it in several easier simpler problems?
    • please please don't re-invent stuff!
    • is it really a problem?
      • ask around
      • you aren't special
      • you aren't that smart
  • Document Behaviour - use the stuff that already exists and make something based on that
    • Procedure
      • collect
      • deconstruct
      • compare

  • OK, so you really have a good idea for a microformat... what to do?


    • Proposal


      • write

      • try

      • feedback

      • iterate


    • Implementation by Scientific Method (as defined in Wikipedia), with a difference - results aren't publish only in the end, but during the whole process




In the end... MicroFormats don't convince me, at all, but the message that you shouldn't re-invent the wheel but yes extend its use is a preety good one.

BTW: graphical card crashed again on this laptop, and thins time I got to see a Kernel Oops. I really really need to do a kernel update here.

ODF: Our Document Future

So, this time it's the ODF presentation, by Donna Benjamin which I already met yesterday.

The presentation talks about three things:

  • Digital Preservation. Why does it matter?

    • Data must be preserved or it is lost. Since nowadays our data is digital, we have to digital preserve it

    • Conversion, Emulation, Migration, Encapsulation...

    • XML is free, readable, open, standard, multi-plataform, and so perfect since it won't age...


  • OpenDocument Format


    • Using an open format is REALLY IMPORTANT for so many reasons that I'll just not going to enumerate here

    • Despite what Donna says, in my oppinion is that, yes, ODF is really great, but it isn't "news"... You have stuff that has exactly the same advantages she said, like LaTeX. The only advantage of ODF to those matters is simply the massive use it has...

    • ODF _is_ the future... but has the concurrent from Micro$oft


  • Australian efforts on the Digital Preservation front

    • Australia's efforts just rulle :-P

    • vers isn't open, but do they know why?




Standardising Web Applications: Rich Web Clients at W3C

Dean Jackson from W3C and he works on Standardisation of Rich Web Clients, SVG, web on mobile phones and such.

He's looking about Standards in the Web 2.0 world. JavaScript applications are using the JS W3C's standard... There are criticisms about W3C not being concerned with the world wide web, but Dean are here to proove it wrong. Increment nowadays standards or create new ones? We don't need "one standard to rule'em all", we can support various approaches, but standards are needed.

Their focus:
* Web API's
* Web Application Formats
* Compund Document Formats

The W3C community is everyone using W3C works, and since everyone uses the web, everyone makes part of it.
* Standardise existing undocumented stuff
* Open channels of communication
* Get implementors involved

We don't have to jump from web 1.5 to 2.0, we don't need to re-invent HTML we're doing another version of it (HTML 5)...

Web API's Working Group:
* Microsoft, Mozilla, Opera, Apple, W3C
* Openness - Live specs available via CVS; all technical discussion is public
* Specs:
1 XMLHttpRequest
2 Window
3 DOM Level 3 Events
4 Selector APIs
- XPath selectors
- CSS selectors (which is going to be released this week)
5 REX
6 Future
- Client-side storage
- File upload
- Network++ (sockets?)
- <canvas> (the element and the API)
* Declarative cs. Programmatic
* Building a Web Operating System

Web Application Formats Working Group:
* HTML working group is needed?
* Web Forms 2.0
- simple extensions to HTML
- significant return for small cost
- some conflicts with XForms
- types like "range", "date", mark a field as "required" (this things are awsome!)
* XBL (they are going to try to adopt Mozilla's work on this)
* Access Control - allowing scripts to access resources from other domains (if I had this while contributing to HCL...)
* Widgets
- Packaging format and conventions
- Metadata for execution
- Media type
- discovery mechanisms
- preety cool to integrate desktop with web, a step to the "Web Operating System" (which I'm not sure I agree with)

Compound Document Formats Working Group - Framework and compounds