Warning: This Penguin Has Teeth!

Written by TAD


This short article is a follow-up to my lengthy moan about Linux and the problems I encountered while trying to install and use it.

I wrote almost 20 articles for Hugi and the suprising thing is how little feedback I got, except for Linux. This article seemed to stir up more emails than anything else, so here goes.

Linux-devotees will be pleased to know that this article will be a little more fluffy than my last rant.


I think Bill Gates is the best thing to happen in the world of computing. He has created one of the largest companies on the planet, is worshipped by millions and builds revolutionary software and operating-systems. His company and employees are pioneers, pushing back the frontiers of home computers and inventing some of the more inventive new features ever to graces our square crt tubes..


In fact the above is what he wants the entire world to believe about himself and his company. I have witnessed many interviews, mostly by the news media circus after allegations of his monopoly, and the more I see, the sicker I feel.

You're jealous because he's rich and you're not.

No, there are plenty of other rich people and companies I respect, but not M1cro$oft. It's their 'messiah' and the other founding ego-maniacs which I really, really dislike.

Looking back to the early, pioneering days of computing you can see why so many people dislike M1cro$oft. I don't want to write a complete history of the home computer and the operating-system we all like to hate, but Gates's company was built on the shoulders of the Xerox corp and all the other visionaries who created the GUI, invented mice, OOPS, CPM and WYSIWYG. The most sickening thing is that Gates and his fellow yes-men completely refuse to admit and credit these people. Take the example of Netscape and its Internet software, did Gates ever give credit to them?

No, never. Instead he tried to force the company out of business. The usual chant he comes out with is: "It's a free market, someone can come along with a better product. They may even become the next M1cro$oft."

Don't be fooled by this 'sportman-like' display. If some competition starts to become successful then you can be sure that the full weight of Gates will start to form a blitz attack plan to either buy out the company, or zero its profits by aping the competition's software and releasing it for free. It wouldn't be bad, except that Gates's software sux.

Hey, I thought this was about Linux!

Well, yes it is. But Linux still has so much catching up to do, that I still think most new users will opt for the 'easy' solution to an operating system, after all, most of the games are only ever released for Windows 95. So most new users will head straight for the games ability of the machine and consider a nice, programming environment much later on, but then, it is usually far too late. The Gates tape-worm is already in place.

The sad fact is that once people are used to one interface they become lazy and resist change, even if that change is for the best. Why do people stick with Windows with all its countless faults?

It's plain and simple. Time.

I am, like most people, guilty of being lazy. Once I have spent the time and effort learning to use an application or operating-system I shy away from trying something completely new. This probably explains why I still use the ancient BRIEF editor (from 1987!) instead of a fancy, new IDE or M1cro$oft Office-block-manager-2000-explorer-system-wizard version 8.

First Impressions count.

So how can people (especially new users) be persuaded to try something new and better, and more importantly, convince them to spare their time and effort and support something new?

The answer is simple.

Make your interface clear, easy and reliable. How many times have you tried to figure out why an option doesn't work, or why nothing happens when you try to perform an illegal operation?

This is my biggest problem with Linux and the installation software which I tried over a month ago. There simply wasn't the user-friendly software which I, like most people, would expect from an OS. If all else fails, have an on-line help (no matter how bad it is). I'm not talking about huge, graphical interfaces, animated wizards or real-time mpegs with pictures of a dork pulling expressions like a lame acting-wannabe, but clear instructions and basic interface/feedback guides.

Before any software should begin to install itself, it should either check for all the neccessary hardware, drivers, libraries, diskspace etc... or just display a list of the requirements so the user can instantly see if they have the correct hardware. There is nothing worse than crawling through a 30 minute installation only to find that a vital driver is missing or that the application needs an extra 50 megs of diskpace, but failed to tell you on the box, before you reformatted your drive. ARRGGGGGHHHHH!!!

Having great software isn't enough. You need to guide new users in smoothly. For example, say you wrote a fantastic new text editor. To newbies it would seem difficult to use and as (Eric said in an email about Linux) it has a steep learning curve. This can be addressed by giving a number of 'ape' options. In the example of a text editor this would be different keyboard layouts (BRIEF, WORDSTAR, etc.). This means that any new user can almost instantly use the software, then over a period of time discover all the new groovy features that their previous text editor didn't have.

Software development should be a development of existing software as well as being a development of the programmer's ability.

Jargon Junkies.

Programmers just love to use strange names and stupid word games to name each item. How many dictionaries are there for computers alone? Hundreds if not thousands. The GUI has to some extent made things easier, because it avoids using names. This not only means less translation is needed for different countries and different languages, but it allows a smoother way to introduce newbies.

There is still no clear naming convention for programs, objects and the interface. Take a sub-directory for example, some call it a 'folder', some 'a directory' and some a 'volume' or 'tome'.

But even the GUI has its problems. That of the artist doing the graphics. How many times have you tried to figure out what an icon was supposed to be? There is some tiny 16x16 pixel block with more colours than a prism inside a 1970's disco fever convention. Again, the designer thought he/she was creating something pretty to use, but in fact the clearity has been lost.

From the screen-shot from the Linux GUI interfaces I have seen, the same problem appears here too. The icons are larger than most people's monitors. And could do with a little TLC on the graphics side (got the hint?). Okay, you get the idea.

You might have noted that I said 'screen-shots' and that's because Linux (Red Hat 5.2) totally failed to find or support my video card, something which Windows 95 also does, but at least with Windows I have some kind of GUI display to reconfigure and locate the correct drivers with. With Linux I was left with a command-line prompt and nothing else, no help, nothing.

A Slice Of Humble Pie?

The feedback from Eric about my, perhaps harsh, treatment of Linux after such a short period of time, was fair and correct. I totally agree that 10 hours is far too short a time to draw an honest conclusion about the OS in general. But I did defend my view that the 'welcome-mat' of Linux (i.e. the installation software) is what most will judge it by, and sadly Linux needs a new, clean welcome-mat otherwise people will go elsewhere.

I would bet that the other people who have dared to say a bad word about Linux have recieved a similiar response from the die-hard Linux fans. But this is identical to the Windows/Moneysoft corp. fanatics. Same knee-jerk reaction, but different OS. I know many people who just love to complain and bitch about computers and software, but the solution should not be an immediate put-down ("stop moaning, loser!") but some time should be spent addressing the problems encountered. In short, if you don't want to get negative feedback to your software, then don't release it, ever!

No software is perfect, each OS and application has its own merits and bugs. But the best software has far more merits and far less bugs than other software. Development doesn't stop when a program is released. It should continue with the benefit of feedback from the millions of users.


The K.I.S.S. principle (Keep It Simple, Stupid!) IMHO is always to best way to design and program software. Like I said in a previous article the GUI is the best interface we currently have, but it still has problems, usually it's the coders doing the designing themselves.

Coders are really good at writing programs for themselves, they know how the final program is going to be used and the kind of task which it needs to perform and those which it does not. They basically know the correct order to do tasks in and if something doesn't happen they know what's going on behind the scenes and how to navigate around the problem. But the users don't know what is happening behind the scenes... and become angry or say bad things about the programmer or company..

The Future..

So what would the next GUI be like and what improvements can be made to Windows, Linux, BeOS and all the other graphical interfaces out there?

Well, there has already been some new ideas demonstrated a number of years ago and this was on national television. The next step is to allow the users to build up their own applications and customize their work environment to the way in which they think and work. I don't mean users having to learn a high-level language before they can type a letter or print their CV.

I believe in the future, Operating Systems and applications will be smaller! Yeah, I know Windows 3000 beta-testers will find this hard to believe and given Mickeysoft's track record it would seem completely false.

Applications will need to share information far more than they do today and perhaps even algorithms will be shared. As electronics become almost sub-atomic it may be that many algorithms will be created in hardware instead of software. It is already happening in the 3d-video card market. Once sprites and line were drawn using the CPU, now dedicated hardware is used. So perhaps most of the common GUI operations will be created in silicon, rather than code. I don't mean just blit-block operations or lines, but entire window structures being passed to some hardware chip which in turn calculates, draws borders, scales bitmaps and handles resizing, dragging objects etc. Given the common features of 3d cards these days perhaps 3 dimensional windows and depth-shading objects which could be zoomed and rotated with very little CPU time.

In short VR on the desktop, or a VR-desktop. It's a fact that a 'large' 1024x768 pixel desktop is still sometimes far too small and representation of data needs the 3rd dimension to help us humans manipulate objects easier.

But isn't all this hardware highly redundant?

Yes, if a new algorithm was developed then the hardware would be outdated. This is one reason why MickeySoft wanted a software Operating-System so it could be updated and upgraded unlike the older ROM based systems. The advantage of a ROM based OS is that it is usually pretty solid with less chance of deleting drivers, or virus attacks. When you consider that people need to upgrade their PC every few years then the software OS argument doesn't look as convincing. Take Windows 95 for example. People with a 486 found that they needed to buy a new machine (ie. Pentium) in order to run the OS. So why not put the OS on ROM like the good old days? It would free up far more RAM, be more reliable and boot a 100 times faster...

The New Kid on the Block.

From the rather small amount of information on the net about BeOS, and some friendly and very informative emails, I already like BeOS.

Athough I haven't actually got my sticky-mits on a copy yet, the signs and views from its users is very positive. I don't mean it's the normal 'BeOS is the king, Windows sucks..." stuff, but descriptions about how stable, (yes, remember that word 'stable') an OS it is. The enthusiasm is apparent from the newsgroups and help I got from many users. The only feedback I got from Linux users was negative, almost attacking me (one of its potention new users!). I'm not phazed by being called a dumb-ass (it's my middle name, grin), but slagging off people isn't likely to motivate me into defending, or developing/buying Linux software in the future, is it?

BeOS post-mortem.

I've had the chance the try out the BeOS demo, even if it only runs from CD and doesn't suppory my video-card, yet. The disappointing thing about the demo was the sheer lack of speed, but this could be due to running off my slow CD drive and only having 44 megs of RAM (the last 4 megs are used by my video-card). So I cannot confirm or deny the reports about its speed over Windows 95, but even with booting from floppy and loading from CD it started up a little quicker.

Whether BeOS turns out to Be the OS of the future remains to be seen. The enthusiasm of its users and newsgroup help certainly convinces me that it's a nicer path to follow than Linux. Overall the BeOS demo ran smoothly and NOTHING crashed, complained or stopped.

I take my Red Hat off the Be. and hope they sort out all of their hardware support. Their GUI looks more 'professional' than the narrow Linux offerings which I have witnessed.

Closing words

One look on the net and you will see countless people trying to create their own OS, just because they've seen BeOS or Linux, or because they are really pissed off at M1cr0soft. Considering the vast amount of time and effort involved in building a working OS, I find it funny that so many 'newbies' post questions like "I'm writing my own OS, it will be free...." and then in the same message "How do I print a character on the screen in 80x86".

Ahh, it's nice to see enthusiasm in newbies... but it's horrible having to break the bad news to them, that they are not "quite" ready to write an OS just yet... perhaps in a week or two (grin).

And I still don't want a fluffy Penguin, Linux leaves me cold. I want something warm, sweet, that runs like honey...... ....and comes from a Be.


TAD #:o)