A Liberty Alliance member said that Passport interoperability is being considered.  I think this is unlikely to actually happen without Microsoft joining LA and LA could break apart if Microsoft joins LA.  This leak is probably just a political move by a faction within LA.

Its amazing how much work each website invests into site navigation.  Tabs, buttons, and menus of all kinds: dropdown, dropside, slideout, slidedown, etc.  This makes no sense from business user's point of view.  Business Browser should sport an Internet Explorer Bar (i.e. Favorites) like view for:

• Authentication Panel (sign-in, sign-out, snowtrack display)
• Optional Shopping Basket Panel
• Site Info Panel
• Hierarchical Sitemap (a tree should do)
• Search Panel

Of course, meme needs a good name so I am going to name it: SiteBar.  I'll try to prototype it this coming weekend.

MasterCard has unveiled MasterCard SecureCode , an implementation of Visa's 3D-Secure plus old initiatives repackaged and realigned into a whole package.  MasterCard's 3D-Secure implementation is an interesting variation in that it uses UCAF.  Just like Visa, MasterCard's press release neglected to mention the role Arcot Systems played.  While this might be acceptable business practice, I dislike such political snubnosing moves transparent to industry insiders.

Snubnosing aside, I like SecureCode better than Verified-by-Visa from marketing point of view.  SecureCode is easy to say yet identifiable and clearly explains it from the customer's point of view: a code used for secure payment.  In contrast, "Verified-by-Visa" is a tongue-twister and explains the technology from Visa's point of view.  Problem gets worse in non-English speaking countries where the task of explaining becomes either impossible with Verified-by-Visa as the marketing flag or fragmented if some country-specific name is used instead.  I recommended SecurePIN before and its not too late to use it.  Similarity with SecureCode will just make it more effective.

While everyone, including Microsoft, is asking the question "What is .NET?", people like me are trying to use it.  As I mentioned before, minimal .NET footprint is large (just like Java) and its runtime are not shared across applications (just like Java).  Memory overhead seems to be 4 to 10 meg per application depending on the range and extent of .NET Framework the application uses.

Three common .NET hosting applications are ASP.NET, Internet Explorer, and the Shell (aka Explorer).  At least with the Shell, .NET runtime is not loaded it is used.  Shell that normally uses around 18 meg of memory ends up with around 24 meg memory foot when one of my shell extensions written in .NET is activated.  I am not too happy with this, of course, but the cost of .NET not too bad for building extensions to IE and Shell.

What does concern me is the increase in memory requirement when .NET applications become common place.  Since each application will need around addition 7 meg, .NET based MS Office suite will cost around 40 megs more if all the MS Office is running.  I usually run around 20 applications and services simultaneously, so around 140 megs of additional memory will be needed if all of them were .NET based.  I expect 512 megs will be minimal memory requirement and 1 gig recommended.  Also, GDI+ is still too slow (its not hardware accelerated) for doing anything fancy.

So we are at a point similar to the time when Windows 1.0 came out and the watershed event, equivalent of Windows 3.1, will not happen for another year or more.

I started using the mouse regularly when Mac first came out (I used mouse before but not regularly) and my fingers were fine for 16 years.  The trouble started when I started using mouses with scroller wheel which is most conveniently accessed using my forefinger, the same finger used for mouse clicks.  After two years of mousewheeling, the tip of my poor forefinger is sore, peeled, and hardening.

So I spent a few minute trying to figure out why mouse wheeling hurts where clicking doesn't.  The answer is that, while clicking is done using the flat area of your fingertip, wheeling uses whole length of the fingertip.  Most common mousewheeling action is scroll down (to read), an action that starts with the flat part of fingertip touching the wheel and ends at the area near the fingernail.  That area is almost never used — poking is not a common activity — so its gets sore really fast.  After two years of it, I have to give my fingertip a 1 minute every 15 minutes or so.

This is why HCI is hard.  A cool new feature could mean either whole lot of evil fingers pointing at you later or our forefinger will evolve to rival the thumb in the future.

HCI stands for Human-Computer Interaction, Computer User Interface in academic words.  HCI is important because it has direct impact on the productivity of computer-based tools.  Poor HCI limits your field of vision and your ability to express.  Font design is part of HCI because a poorly designed font will surely limit the time you spend staring at text rendered using that font.

Recent mock up of digital dashboard by Rajesh Jain and Ray Ozzie's comments is inline with what I am working on: Groupware HCI.  Toughest part of designing groupware is its UI because the whole aim of groupware is to enhance group productivity and, as I mentioned above, HCI has direct impact on productivity.

Rajesh's web-based dashboard fails because

• power of dashboard UI lies with its realtime view and web pages are just not responsive enough no matter how much JavaScript you throw at it.
• a dashboard is more than just a collection of unrelated Post-It sized web pages.  Position of and relationship between each panel is critical to a dashboard's effectiveness.
• web pages consistently convey the feeling of Places and as web browsers convey the feeling of Movement between Places.  This conflicts with the Personal Gadget message conveyed by dashboards.

Fonts are one of those things that you work with everyday but don't truely appreciate until you start running into some bad fonts.  Courier New, for example, has these casual and consistent curves that enhances readability without poking your eyeballs.

Because I have a font-fetish, I hate programming in ugly jagged fonts.  Unfortunately, there aren't many fonts designed with programmers in mind.  These days, I use 12 point Courier New font on my XP running at 1600×1200 with ClearType.  Now even spagetti code looks great.  I am not completely happy, of course, because fonts for programmers need to be more compact vertically so I can see more lines of source code at one time.  At one point, I even designed my own called Codex but they didn't scale nor print too well.  Amatur Fontographer!

I came across a new font today called Anonymous, but its got some really annoying curves.  I thought I would mention it just in case someone has different taste than I do.  What I would really kill for is a professionally designed font designed for programmers.

Open source, as a meme, is a successful one.  It is also a confusing one because open source often means free software.  I am for open source, but I am against wholesale adoption of free software because I believe it harms the health of software industry.

I typically write free open source software and tools for bleeding edge technologies.  I have written one of the first implementation of W3C DOM API while the spec was still being worked on.  I did this for the XML community so they could start using the API early and provide feedback to the DOM WG.  I have stopped working on my DOM implmentation after many DOM implementations became available.  I felt it was time to encourage commercial development.

Well, the XML tools market didn't work out as I hoped.  With continued flood of free open source (often Java) tools, XML tools market is mostly dead.  Why should anyone pay for XML parsers, editors, databases, and servers?

Free open source software devalue commercial software and poisons the marketplace.  If some talented individual started giving out free Microsoft Word clone with source code, not only will Microsoft stock drop by 25%, entire word processing market will disappear.  Once buys starts thinking that something should be free, there is no turning back.

One might argue that adding new features could revive markets decimated by free software.  New features has leverage only during early half of a product category's lifecycle.  Word processing market is already well past the halfway point.  I measure halfway point as the point where 20% of product feature set meets 80% of user's needs.  Past that point, users start caring less about new features.

In the old days (10-20 years ago), I used to frequently visit bookstores and computer stores just to see whats new.  Since I am an avid reader, I would typically end up buying a stack of books.  With software, I was more frugal because, being a programmer myself, I was more critical (why buy what I can build?)

These days, I hardly visit the bookstores nor computer stores because I have no need to.  With so much information freely available through Google, I don't see the need nor time to read books (I do occasionally read good books like Tipping Point and Harry Potters :-).  Besides, I ran out of places to put all my books long time ago.

So, I am wondering how the bookstores and computer stores manages to stay in business.  Is it just me or is it the coffee and cables?

OpenEvidence is an EU sponsored open source project to build core digital evidence technologies.

OpenEvidence produces technology for "evidence" creation and validation of electronic documents, meaning "evidence" a document certified by some authority that guarantees the data it contains. The technology developed by the project can be used as basic building blocks to support such services as non-repudiation of electronic business transactions, property right protection and notarisation.

Frankly, I think they are digging a big hole without a way out.  Security means nothing if its not usable and EU laws tend to ignore usability.