Blunder Dome Sighting  

Hangout for experimental confirmation and demonstration of software, computing, and networking. The exercises don't always work out. The professor is a bumbler and the laboratory assistant is a skanky dufus.

Click for Blog Feed
Blog Feed

Recent Items
nfoWorks: IS 29500 (OOXML) Moving Ahead?
Cybersmith: Resources for .NET Interop
A Conceptual Architect Am I?
Cybersmith: Attributions for Code You Use
DMware: ODMA's Dark Matter
Cybersmith: No-friction Bits and Pieces
VC++ Novice: Is Native C++ a Dead Language?
nfoWorks: What Are those Harmony Principles, Agai...
nfoWorks: Tracking OOXML DIS 29500 into the Blue
nfoWorks: The ISO/IEC Harmonization Opportunity

This page is powered by Blogger. Isn't yours?

Locations of visitors to this site
visits to Orcmid's Lair pages

The nfoCentrale Blog Conclave
Millennia Antica: The Kiln Sitter's Diary
nfoWorks: Pursuing Harmony
Numbering Peano
Orcmid's Lair
Orcmid's Live Hideout
Prof. von Clueless in the Blunder Dome
Spanner Wingnut's Muddleware Lab (experimental)

nfoCentrale Associated Sites
DMA: The Document Management Alliance
DMware: Document Management Interoperability Exchange
Millennia Antica Pottery
The Miser Project
nfoCentrale: the Anchor Site
nfoWare: Information Processing Technology
nfoWorks: Tools for Document Interoperability
NuovoDoc: Design for Document System Interoperability
ODMA Interoperability Exchange
Orcmid's Lair
TROST: Open-System Trustworthiness



Cybersmith: The Confirmability of Confirmable Experience

Finding ways for the experience of users to be confirmable by the producers of software is increasingly difficult as we operate with distributed applications over networks and the world-wide web.  Because we can't directly show another user or the software producer what our experience is, we need forensic tools that allow us to capture and communicate the locally-observed behavior to others who are elsewhere.   I always keep screen capture software handy.  An experience with the new Internet Explorer 8 beta 2 release demonstrates the value of that.

Screen Capture: the Primo Confirmability Utility

One of the most-important tools for cybersmiths, including power users, is a screen-capture utility.  Whenever I set up a new computer, my favorite screen capture utility (currently HyperSnap 6.30) is one of the first two products I install.  (The other is WinZip for its value in addition to the built-in Zip capability of Windows Explorer.  That's actually in a three-way tie with my password-safe utility and Microsoft OneCare.)

If I could count on a screen saver being available during initial set-up (even log-on if that were possible) and configuration of a new computer's operating system, I would be even happier.  I want a screen-shot record of everything that I go through and of every option and setting and parameter that I choose.  I do the same thing whenever I am installing a new software package for the first few times.

And whenever there is an unusual incident, I start grabbing screen shots as long as I am able.  If I can't make screen captures, I will grab my digital camera or (though needing to get the hang of it still) my Windows Mobile cellular phone.

Although there is limited screen capture capability built into systems like Windows, I rarely want the entire screen.  Also, I want to save in a loss-less compact format, almost always preferring PNG format.  This format is easily included in e-mails and posted on a web site to back up an incident report or provide documentation of something interesting.  All of the VC++ Novice screen shots have been created this way.

Screen Capture: Do You See What I See?

In today's world of distributed applications, one of the greatest difficulties is dealing with interoperability problems (e.g., garbled e-mail messages, broken web-page presentations, and document viewing/presentation glitches).  Even if an offending file or document is sent back to the source with an incident description, the recipient may not see what you saw.  Reproduction isn't even the first problem.  Clearly seeing what it is you experienced is the first problem.   There needs to be an out-of-the-failing-channel way to deliver a visible rendition of what you are seeing.  Sending a screen shot will do that. 

Confirmable Experience for Trustworthiness

This ability to report your experience in a way that a distant party can confirm it is a critical need in today's richly-variable and richly-connected world.  I expect that we will eventually rate software and its support by how well it lends itself to confirmation of user experiences, along with easy reproducibility and remedy of unexpected situations and usability difficulties.  The first step is to be able to demonstrate what happened in a simple way.  Today, use of screen shots is one of the easiest and reliable ways to do that.  There often needs to be an accompanying narrative and any data or files involved, but the screen captures are essential for clarity.

My colleague Bill Anderson and I use the term "confirmable experience" almost habitually when discussing interoperability breakdowns, inscrutable installation instructions, and the occasional flagrant exhibition of system incoherence.

There's another kind of confirmable experience, and it is the one I can avail myself of as a developer and software producer.  There are many cases where I need to find ways to confirm what kind of experience my software products (including web pages) are providing and what the breakdowns are.   Screen capture is useful for documentation and demonstrations as well as being an important forensic tool.

An Example with Internet Explorer 8.0

I am one of those people who provides an absolute minimum of testing for web sites and blog pages.  I figure that if the material shows properly in my authoring tools and also in my usual browser (yes, Internet Explorer), I am on safe ground.  Also, because I do not use scripts and I have static pages almost entirely (with the occasional embedded video frame), I rely on pretty-simple HTML. 

I also figure that I am fairly safe from problems with standards-strict browsers of various flavors.  I make sure that my recent web pages are in UTF-8 (to minimize character encoding issues) and have an unassuming DOCTYPE declaration (for HTML 4.01 transitional mainly).

I've never received a report about rendering difficulties with my web pages, although that's not a reliable test.

Knowing that Internet Explorer 8.0 is going to usher in a period of widespread standards-strictness, I figured that I was still relatively safe.  Just the same, when IE 8.0 beta 2 was released the other day, I decided it was time that I took my chances with the browser in order to find out how well my sites will be viewed (expecting an immediate "pass" grade of course).  Another value of confirmable experiences is separation of fantasy and magical thinking from reality.  Sure enough ...

Once I had installed IE8 beta 2, my first check was with my default home page,, where I regularly check the weather, headlines and movie listings for my favorite local cinema.  The page passed the eyeball test, although I now see that there is a broken-compatibility indicator compatibility-view button.  The indicator really does disappear when a page is "clean," specifically selected in compatability view or standards-mode, as I can see on some other pages.  If there is an indicator and I push it in (it is a push-on, push-again-for-off button), I can see minor changes on that home page.  This is exactly what I want for a site-forensic tool, and IE8 makes it easy for me to review my site, find the discrepancies and, if serious enough, repair them.  There is also a third case, where I add metadata to pages or an entire site so they will be automatically-recognized as non-standard under compatibility view and treated in a forgiving way by IE8 the same way its predecessor had by default.  Finally, users of IE8 can select compatibility with loosely-standard pages themselves, and IE8 will remember that selection on a page-by-page domain basis.

Illustration of IE8 beta 2 Compatibility Control: (a, left) strict presentation with compatibility defects, (b, center) rendering with relaxed rules, (c, right) absence of indicator with fully-standard page

(a) IE8 strict interpretation with compatibility indicator (click for larger image)

(b) When compatibility is selected, IE8 provides loose interpretation (click for larger image)

(c) IE8 recognition of strictly-compliant pages has no indicator (click for larger image)

Notice, also, that the experience I had is with my, not yours or anyone else's.  This is an easy way to demonstrate to someone what my experience is, no matter what their own checking reveals.  I love how this is working out.  And of course, my simpler pages would fare well.  Sure ...

Not So Fast, Sparky

One of the oldest web pages that I still have on a site is the home page for Orcmid's Liar,  This is a trivial page, essentially a place-holder for riches yet to come, while I distract myself with blogging, other content, and little concern for this site's puny entrance.  I expected this to be a no-brainer.

Whoops: Presentation of my home page in IE8 beta2 strict-standards mode is unbelievably ugly.
Strict-standards rendering is terrible and incompatibility is indicated (click for full-size image)

The result is terrible.  The single-row, three-column table at the top of the page is mangled.  There is text missing from the central cell and the table does not span across the width of the page as intended.  It looks like the table has been wrapped on the page with the third cell below the first two.

Compatibility mode (that is, loose enforcement of standards with a best-efforts rendering) provides what I am accustomed too:

Expected result: Viewing under the forgiving compatibility mode provides the expected rendition.
Compatibility mode of IE8 beta2 provides the expected view (click for full-size image)

This is a common page top-heading format used throughout my web sites.  I nosed around to see if they all render so badly in IE8 strict standards mode.  There are incompatibilities, but not the one seen on the Orcmid's Lair home page. 

While wandering around, I noticed another problem.  Sometimes when I return to the home page, viewed in strict standards mode, by back-paging from a different page, I see a rather different incorrect version.  This one is peculiarly almost-right:

Erratic strictness: If the page is visited, navigated through, and then back-paged back, the strict rendering is different.
Revisiting the defective page by paging-back from another site page almost has the expected rendering (click for full-size version)

Exactly Whose Experience Are We Experiencing?

Granted, my home page is not strictly-standard according to IE8 beta2, and it is presented properly only with the forgiving compatibility mode for loosely-standard HTML.  At the same time, absence of consistent behavior is an indication that there may still be defects in the IE8 implementation of strict-presentation in accordance with web standards.

I was still left with the problem of straightening this out.  It seemed to me that the page is simple enough that I could repair the page and not resort to the special meta-tag to request compatibility mode in browsers, such as IE8, that will recognize and automatically accept loosely-standard pages and render them as well as IE7 does.

I attempted to edit the HTML, making adjustments to the 3-column table that I though might help.  There was no change.  But there are other pages on the site with the same 3-piece top heading and those render fine.  So I copied one of the successful 3-column tables in place of the current one, edited the text appropriately, and found success:

That's more like it: The repaired table, shown in this narrower view, is presented just fine in IE8 beta2's strict-standards processing.
Repaired welcome page renders the title properly although there are apparently minor incompatibilities somewhere on the page (click for full-size version)

The compatibility button has not disappeared, and there is more to accomplish.  But the ugliest part has been repaired successfully.  I'm not sure what the difference was, so there is more to learn.  

The next action, for me, is to use an HTML validator that lets me attest to the validity of the pages on my sites.  That will come later as I find occasion to review all of the pages, treating the most-embarrassing glitches first.

I also must deal with the fact that all I have done so far is treat symptoms, without a clue regarding the underlying cause.  This is particularly bothersome because the cause might be the beta2 status of IE8 and not entirely a matter of a page's non-standardness.

This post has more confirmable-experience dimensions than revealed by the immediate content:

  1. It is a prospect for confirmation of standards-compliant presentation via IE8 beta 2.
  2. It is my first use of Blogger Labels (Categories in other blogging systems) to archive posts by categories as well as having weekly chronological archive pages and the buckets of individual posts made in the same month.
  3. I am using Windows Live Writer image-uploading capability to FTP images to a directory of the blog and this is the maiden exercise to confirm that is functioning properly.

All of those arrangements are being explored with this single post.  There's a principle about making only one change at a time that I am neglecting.  OK, I feel lucky.  This is the Blunder Dome, after all.

Update 2008-08-30T16:43Z I made some awful errors ("two-days" instead of "todays") and chose to tweak the page.  I also notice that there are more problems that I don't know what to do with.  It is time to come up with an IE8 mitigation approach.  Stay tuned.

Update 2008-08-30T18:24Z I did achieve the three tangential objectives.  As a result of that and the experience reworking the home page, I am adding an IE8.0 mitigation category for this post and others to follow on this and other blogs of mine.

Update 2008-08-31T02:46Z After developing further analysis and checking out further resources from Microsoft, I realized that I misunderstood the significance of the Compatibility View and what has the button be present.  This page has been touched up accordingly.

Labels: , , , , ,

Construction Structure (Hard Hat Area) You are navigating Orcmid's Lair.

template created 2004-06-17-20:01 -0700 (pdt) by orcmid
$$Author: Orcmid $
$$Date: 10-04-30 22:33 $
$$Revision: 21 $