Introduction About Site Map

RSS 2 Feed RSS 2 Feed

Main Page | Blog Index

Archive for November, 2005

Research Presentation

Poor appearance model
An appearance model of the brain after
perturbation has been applied to its training set

LATER today I will deliver an important presenation on my recent research work. This work has been done in collaboration with UCL for the past few months. This comes amidst 2 days of talks (the first day was yesterday), which slow down my activity on the Internet.

The presentation file that I practice with is in OpenOffice 1 format. This relates to my discussion on one-file presentations versus the more open and sparse (and in my opinion correct) form, as argued and published a couple of days ago. In-depth details on academic progress are periodically posted in MARS, which is a separate, research-related section that jointly resides on this domain.

Disinformation as a Windows Reinforcer

Bill Gates
Bill Gates arrested in his younger days (photo in public domain)

From previously-cited writer comes yet another ‘killer essay’.

if (Windows Rules) then (Linux fails)

Part of the problem with the documentation and identification issues I talked about last week – and will talk about more later – is that it is very hard to separate information from disinformation.

Disinformation comes in three major forms:

  1. innocent mistakes;
  2. intentional disinformation (aka FUD); and,
  3. (self) delusion.


Read on…

Also see: Linux is not Windows

RAID Redundancy

Servers stack

RAID stands for Redundant Array of Independent (or Inexpensive) Disks. As yet, I find RAID somewhat redundant, at least in my individual scenarios and for my personal purposes.

RAID is associated with backups on multiple disks, which in turn enables easy data recovery. The disks may also be distributed across geographically isolated and thus independent places. RAID does not directly cater for synchronation of data, applications, settings. That, for instance, is one of the powers of on-line services and network-hosted environments. This, as a matter of fact, is what pushes computers towards becoming relics.

One question to consider in RAID is the target of a write/read operations. This point refers to where data should be fetched from or save onto and how changes should propagated among the multiple disks. These issues are easily resolved in the following way:

Let us look at this conundrum, which is flawless management of files across computers. The confusion over where data gets stored and fetched from can be resolved owing to remote login. Assuming we speak of high-bandwidth networks here (100Mbit LAN in my case), the ‘cost’ of using SSH bandwidth is almost negligible. Even 10Mbit connections would give you a very responsive UI behaviour over the network. I know this because I tried it. The exception are cases where you stream many large frames over the network (e.g. video, games).

I have never looked deep into RAID technologies. Why not just scp (or rsync) the entire content of a hard-drive periodically? I maintain files on just a single machine, which I always SSH to from elsewhere. Its contents get mirrored on two other remote machines. Files that change on a daily basis are backed on the SAN overnight using cron jobs. This simplifies life a great deal. I still see no compelling reason for using RAID, especially with the existence high-speed networks.

As for Web site backups, these can conveniently be downloaded unless an automated job is set up, which is something I hope to do one day.

Related short items:

Grim Outlook for the Media

Man and his dogIn this era we live in, there is a mounting fear among journalists. No longer is it only books and papers that are able to reach people’s attention. Nowadays, people can get delivered content that they most desire by using search engines. People also discover ways of getting high-quality, credible information without paying a penny.

Realising that adaptation to change is needed — somewhat of a new status-quo — newspapers begin to penetrate the Internet. All the same, they fail to make profit that they have become accustomed to over the decades. Sooner or later, methods similar to eavesdropping are employed. Yet again, the grim outlook for the press seems inevitable and ever irreversible.

Rupert Murdoch has forecast a gloomy future for newspapers with the growth of the internet, saying he doesn’t know “anybody under the age of 30 who has ever looked at a classified ad”.

The owner of the Sun, Times, Sunday Times and the News of the World, who once described newspaper classified advertising revenue as providing “rivers of gold”, now says: “Sometimes rivers dry up”.

The media is not alone in this. The Royal Society attempts to keep science off the Web, arguing it could harm scientific debate.

The Royal Society fears it could lead to the demise of journals published by not-for-profit societies, which put out about a third of all journals. “Funders should remember that the primary aims should be to improve the exchange of knowledge between researchers and wider society,” The Royal Society said.

Multi-file and Single-file Presentations

MIAS IRC presentation
An example multi-file, Web-based presentation (click to start)

RENCENTLY, I have being passively swayed (“forced” rather) towards using WYSIWYG applications for composition of a presentation. I have done that as a teenager, but I sure know the issues associated with the paradigm. I use OpenOffice at the moment, in conjuction with colleagues who stubbornly stick to PowerPoint.

As sad as this may sound, it seems obvious that, judging by the large, conferences do not accept presentations in formats such as HTML, even if they are as advanced and rich as S5 (exemplified above). Many conference them make PowerPoint the sole option, thus snubbing anybody who does not use Windows or is unwilling to invest in expensive licences that lead to vendor lock-in. Fortunately, I work in O/S-tolerant environments, so I rarely need to suffer from such narrow-mindedness.

WYSIWYG is a good paradigm in principle. Alas, as explained several times in the past (e.g. Dangers of Abstraction), WYSIWYG tends to be lossy. Also, re-use becomes as serious problem; searching and indexing likewise. The idea of encapsulating objects like arrows and circles in a single file, along with videos, images and text is ‘unhealthy’ to say the least. A binary presentation is not one which is open for standards-based tools to interpret. It is also poorly-structured.

OpenDocument does not make it trivial for one to extract individual objects unless various tools are used. Therefore, I am reluctant to ever use OpenOffice and yet I recommend and promote it. Hypocrisy? I say “nay”. Its entry barrier is low, and if not the cost, then open formats should make the difference and end the vendor/product dependency. OpenOffice will never suffer from the same pitfalls Microsoft vainly ignore.

An HTML presentation has its text stored separately (under e.g. index.html). Additionally, there are videos and graphics as individual files in their own isolated directory. The technology which drives S5, as in this particular case, involves a collection of stylesheets which are easily exchangeable, interchangeable, and are stored aside, together with their associated graphics. Lastly, there is JavaScript to integrates this technology and make keyboard navigation, buttons and the like a practical reality.

To finish off with a rant, why is it that ‘simplicity’ ended up attaining a huge tar-like file which is obscure in terms of its content? If it were not for Microsoft Office, would somebody else have done the same? Some time ago I heard that Office files were bound to become a compressed files which contain all the peripheral files, e.g. images and graphs, completely apart. XML is frequently echoed by the media in this context. Could this be related to Microsoft’s recent proposal to open up their formats?

Spherical Display

Spherical monitorI happened to find these astounding monitors, which reminded me of spherical desktops. Put in the words of the vendor: “The Elumens VisionStation allows for a fully immersive display of 160°. The VisionStation’s ultra-wide FOV creates an amazing sense of space and depth, without need for goggles or glasses“. All the same, such a display would of course be quite pricy. It doesn’t appear to offer substantial and pragmatic benefits unless one uses simulators.

Also see: Another odd use of a dish

Name Collision

Car crash

When names collide

BACK in 2002, I chose to work on a project which dealt with the game Othello, also known as Reversi. To put rigour into development, I chose a name for it. I did so without paying much thought to any future potential. A careful and exhaustive investigation of name collisions simply did not seem worthwhile at the time.

Having searched the Web at a shallow level, I did not know of any name collisions when I chose the title “Othello Master”. I even explained about the choice of the name in my report and proposal. It refelects on the way I viewed the choice of the name at that time:

This project has been set to produce an application which will be titled Othello Master due to some visual similarity to an older game called Chess Master. It will require knowledge of game theory and advanced computer graphics.

The name was therefore conceived in a most innocent way. It was only less than a year ago that I became aware of a name collision, for which I am to blame. Search engines had revealed a game from the mid-eighties, which suddenly resurfaced in results from archival pages. It ran on the Amiga, but perhaps on other platforms too.

I sometimes wonder if I should get a hold of this game and play it. Mine is Open Source and GPL‘d so no-one is prevented from playing it for free. In fact, the downloads page is always there for those interested. As for the number of downloads, I believe it itches 1,000, but I rarely keep track of the numbers. It can run on all platforms and there is even a Windows executable.

Other items on Othello Master:

Retrieval statistics: 18 queries taking a total of 0.180 seconds • Please report low bandwidth using the feedback form
Original styles created by Ian Main (all acknowledgements) • PHP scripts and styles later modified by Roy Schestowitz • Help yourself to a GPL'd copy
|— Proudly powered by W o r d P r e s s — based on a heavily-hacked version 1.2.1 (Mingus) installation —|