Category: Architecture

Basic features that are missing in browsers

With advent of web2.0, the software world has rapidly changed. Native apps have become obsolete and everything works using a web interface. In fact, some computers (like Chrome OS based) are being designed purely for web browsing. The same is true for mobile and tablet platforms, the browser being a significant player with reports suggesting uses spend an average of 25% of their time on browsers.

There are a few features that are still missing browsers:

– Switching: Normal apps can be easily switched using Ctrl-Tab

– Auto association of file types

Key board loggers – why not? Conceptually they can be made secure.

Every now and then, comes a time when you want to tear your hair off because you typed (and clicked) on a bunch of info into a form and now you don’t have access to it. You want a time machine. Just to go back a few mins or hours and get back all that data.

Why don’t we have technology to make that happen?

A keyboard + click logger would do the job. These programs exist, yes, very much, but are touted as insecure. Many computer worms, virii have key stroke loggers built in them and in those cases their intent is obviously malicious.

But why can’t we have an ethical, secure keyboard logger? log the keystokes but then encrypt them and keep. Make the data accessible only after two factor authentication. The security issue can surely be solved, can’t it?

If scope of capture is a problem, then let’s focus and apply solutions to that problem instead of just branding the whole thing as unsafe. We could easily have loggers that ONLY capture browser based interactions.



Planned Obsolescence for IT Architecture/Design

One of the common dilemmas that IT Architects face is of long-term vs. short-term solutions. Finding the right mix is more an art than trade. Can IT Architects take inspiration from other disciplines?

The answer is a resounding yes. Although for completely different (and often mean profit oriented reasons) the electronic industry implements ‘planned obsolescence’. Chip manufacturers have been known (infamously) to have devices have higher rates of failure after a certain period of time. The wry with mystic reports of phones starting to fail near the telecom contract providers contract expiry date.

Can’t we implement ticking alarms in our code and/or design so that the system start to fail (or call for attention) if a long term solution does not replace a short term solution within a stipulated amount of time?