February 25, 2004
There is nothing wrong with optimization per se. Incremental improvements are almost always welcome. But there is something wrong when we spend an excessive amount of our time and resources on the incremental improvements that come from optimization while missing out on the kinds of radical improvements that can spring from innovation.
Sure, there are times when a focus on optimization is "better" than innovation, such as when it would be too disruptive to interrupt the operation of an existing process, where people are very dependent on not rocking the boat. Incremental and evolutionary improvements can frequently be made with minimal disruption.
But, there are frequently times where the modest benefits from optimization simply aren't good enough. Sometimes band aids are insufficient to achieve the necessary level of improvement and radical surgery may be needed. Sometimes people are so disappointed by the status quo that the kind of revolutionary improvements that can come from true innovation are preferable, no matter how disruptive they may be in the near term.
My suggestion: We need to focus more of our resources and talent on radical, revolutionary, earthshaking, disruptive innovations and defer more modest innovations that frequently consume more resources than they save. Let the chips fall where they may. In the long term, we'll be more satisfied with the results.
February 15, 2004
The advent of instant messaging was hailed as a breakthrough, a tool for enhanced communication. But this bias towards the merit of real-time communication doesn't do justice to the merit of the considered thought and reflection that characterize asynchronous communication. Sure, real-time conversation does have value, but can also be abused. Besides, we already have the telephone with relatively cheap connections, so why all the exuberance?
The real danger of IM is that the irrational attention distracts from the importance of continuing to enhance the asynchronous capabilities of email.
In fact, maybe all IM really needs to be be blended with email so that the two are seamless. Imagine sending an email with an "invite" tag that will alert the recipient that they can opt to convert the email exchange to an IM exchange. There are a lot of interesting possibilities that can't even be considered as long as the world of IM is considered wholly separate and "better" than email.
There are times when people email back and forth in almost real time and it would be nice if the IM/email software could detect that and offer either party to switch to IM mode, and later possibly back.
The real, cosmic significance of email is the asynchronous mode which allows parties to communication without needing to be online at the same time. It's infinitely more convenient to send an email than to play phone tag. And if the sender is forced and given an opportunity to edit and review their message before sending, there is a much better chance that the recipient will have an interaction that is both more succinct and much more economical.
February 10, 2004
We spend inordinate amounts of effort attempting to design complex structures for representing and processing information. The problem is we that we frequently guess wrong about the most appropriate structure or even if we have picked the best structure, something changes in the real world and all of a sudden the old structure is no longer "best".
What is important is that we design our structures so that they can be readily morphed into new structures as the need arises.
Sometimes that means designing extra redundancy into the structures so that they can be morphed more easily.
The best approach is to use a keyword-oriented relational approach so that the actual data can be stored in a robust canonical form and then morphed into temporary structures as required. Besides, it never hurts to have the "master" data stored in a remote, safe location anyway.
January 28, 2004
There is one question that has bugged me since day one of "the internet"... Given the military origins of "the net", WHY ON EARTH were the protocols DESIGNED to be insecure?? Given military paranoia and general popular disdain for eavesdropping, wire taps, snooping, etc. even back in the 60's and 70's, how could ANYBODY have possibly IMAGINED that a network, including email and web pages, that was insecure BY DEFAULT was even remotely acceptable??? Even if Tim Berners-Lee had some kind of European agenda for "openness" (ala the old MIT ITS "gentlemen's time-sharing") for HTTP, why were the underlying TCP/IP protocols DESIGNED to be insecure??? Can anybody point me to a URL that gives a SOLID justification for foisting such an insecure communications model on an unsuspecting world? Or was it simply a conscious decision by the designers of "the net" to "build" an environment in which many, many consultants and add-on security providers could flourish?
November 18, 2003
Whereas a massively parallel computer may have thousands or even tens of thousands of processors, an ultra-massively parallel computer would have millions of processors. One example might be a digital camera or video camera in which a processor is available for each pixel. Another example might be a computer display with one processor per pixel. Another example might be an airplane with a processor for each element of the surface and airframe to dynamically adjust the shape and flight characteristics of the plane. Another example might be a large scale simulator to do fine-grained simulation of real-world systems.
August 3, 2003
Everywhere we look we see, hear, feel, and otherwise sense technology. Some people seem to revel in the artifacts of technology (the gadgets), while others are merely annoyed. My idea is that the goal is twofold: 1) technology should be everywhere (ubiquitous), and 2) technology should be invisible (transparency), not directly sensable. This approach is clearly not a priority criteria for most designers. For example, web sites, personal computers, PDAs, and cell phones are getting more complicated by the day rather than simpler and more powerful (measuring 'power' by the criteria of how much the technology does for you without effort on your part.)
Embedded technology (appliances, motor vehicles) do seem to be taking some of this to heart, but seemingly by accident (and limitations of cost to provide display and user interface technology artifacts) than by intentional design.
Our success with technology tomorrow should be measured far less by the wow factor of the artifacts than the fact that people can make indirect use of the technology without even giving a second thought to the possibility that technology is involved in the everyday tasks of life.
July 3, 2003
I hate URL's. Even with my software development background, I find the crazy syntax and structure of web URL's to be truly offensive and certainly not user-friendly in any way. My big complaint though is that they are a truly lousy way to do bookmarks. I'm writing this rant right now because for the umpteenth time I have clicked on somebody's external link and found that it is broken because the external site has shuffled their pages around. So, the heart of the problem is that URL's explicitly code the structure of the web site rather than using some sort of logical syntax. In this most recent case, I'm looking for the "glossary" page of the site and the web-master of that site, in his infinite wisdom, had some reason to change the folder names or structure, rendering any bookmarks to the glossary completely broken. What is really needed is a keyword-oriented mechanism for drilling down into a web site, so you would specify the domain name, the keyword, and possibly some arguments. In fact, there is absolutely no good technical reason for any user to see URL's in their browser address bar or bookmarks that have the explicit folder names, structure, or even detailed page names (let alone all the oddball file extensions that get used.
And on a final note, there should be a way for a program (such as a browser) to query the keyword list for a site (which may have a hierarchical structure) so that site maps can be displayed in a standardized (but user-controllable) format.
June 25, 2003
What idiot decided to make a capital "eye" (I) look virtually identical to a lower-case "el" (l) in san serif fonts (as in "Il" -- that's a capital "eye" followed by a lower-case "el"!)?? I write this because I was just now scratching my head wondering why one of my anti-spam rules wasn't working. I had an email filter rule that catches "inkjet" in the subject line and it wasn't working. The offending email had "lnkjet", spelled with a lower-case "el". Alright, I'll award some bonus points to the spam hacker, but I'll take quadruple that many demerit points from the font designer(s). Personally, I prefer serif fonts. Besides, I personally have had legitimate visual problems with san serif fonts, especially where some data needs to be entered in case-sensitive format.
This problem is so bad that in Washington, D.C., which has lettered streets including "I" street, people have taken to referring to the street as "Eye Street". In fact, at least one building sign uses that name. It's also popular on business cards.
Of course, there is also the problem of the letter "I" being confused with the roman numeral "I".
June 21, 2003
Put simply, they don't get it! Apple is marching to their own drummer, and he doesn't seem interested in having me as a customer. I would willingly give up my PC and switch to a Mac, but Apple has made it too difficult and expensive. My complaints about apple:
June 21, 2003
Dell has a tradition of promoting a low price, but for a truly lame configuration that must be seriously upgraded to make it barely acceptable, including processor, memory, disk, CD/DVD, and screen. At least in the past, all those upgrades to get to simply a "standard" configuration from other venders resulted in a more expensive notebook computer than available from Toshiba or HP. They do appear to be getting better, but it still really annoying to have to do all that customizing to get to a reasonable 'base' configuration.
June 21, 2003
Although significant advances have been made in LCD screens over the past few years, the average LCD screen (now a 15-inch XGA) still suffers from the "screen-door effect", with fairly bright pixels meshed with the black space between pixels. In addition to being annoying, this is bad for your eyes as well, and noticeably slows the pace at which your brain can read and assimilate information from the screen. My solution: a quantum jump in pixel density so that pixels disappear. In other words, a high enough density so that your untrained eye simply will no longer see the boundaries between pixels. This is somewhat analagous to what happened when we moved from dot-matrix printers to 300 dpi laser printers and finally to 600 dpi laser printers and beyond. The leap from 300 dpi to 600 dpi plus some 'tricks' is what did the trick. A typical LCD screen is barely 100 dpi, so we have quite a ways to go.
We'll also need to make a quantum leap in software since we still have way too much pixel-oriented 'bitmapped' images, such as icons, window border effects, and small fonts 'optimized' for the approximate physical size of a pixel. Anybody who has seen a UXGA screen without 'tweaked settings' knows the problem. There is no excuse (other than simple technical incompetence) for the inability of current software to cope with varying pixel sizes. In fact, Adobe PostScript and PDF handle higher pixel density very nicely. One of my personal needs for higher resolution screens is to facilitate the reading of PDF documents.
June 20, 2003
I can't even begin to estimate the amount of time, resources, and user confusion that has been consumed by computer software that depends of 'case sensitivity'. Computers are supposed to be able to automate tedious and error-prone details, and getting case right is one of those tasks. I'm writing this rant right now because I once again visited a web site that felt obligated to clutter up the screen and my consciousness with an imperative that I, the user, need to worry about case sensitivity. It was a logon screen and said "Note: Both your LoginID and Password are case sensitive." Note that training and customer service resources will be consumed by people who accidentally got the case 'wrong'.
Quite a number of systems designers have made the technically correct choice and eliminated the case sensitivity issue, at no net loss of functionality to the user or the underlying software. It is beyond my comprehension why not all designers and developers get this seemingly simple issue right.
I guess the problem is that you can't legislate against stupidity!
[ Home | Books | Travel | Contact ]
Updated: January 30, 2006 08:45:02 PM -0500
Copyright © 2004 John W. Krupansky d/b/a Base Technology