Saturday, June 18, 2005

Nicolas Roard's amazing StepTalk palette

As the author/maintainer of Gorm this was actually something I had meant to do myself, but Nicolas has done it!! It's amazing. To be able to edit and modify StepTalk scripts and virtually create the entire application within Gorm itself is quite something.

Please see the "Random Thoughts of a Camaelon" link.

Friday, June 17, 2005

The Book of Rules

Here's my book of rules. Occasionally I quote this. It's a collection of personal facts/rules that I have been compiling for some time.

Here it is:

Greg's Rules

0. In any situation involving people, there will always be politics.

1. 99% of the general public ( including technical people ) are not willing to learn anything new unless they absolutely must.

2. Most people are out for their own interests only and do not care about the big picture.

3. The mainstream media is both fickle and stupid.

4. Given a set of technologies, the worst one of the group will almost always prevail since it is
understandable to the lowest common denominator of people (see rule #1).

5. The general public will crucify that which doesn't follow the norm (again see rule #1).

6. Being a realist about how software development works is essential, idealists tend to always loose.

7. The next guy who stands up and says "look at me i've got all of the answers", usually doesn't.

8. Most things which start with Open* usually aren't. Examples OpenLook (proprietary), OpenVMS (proprietary)... etc.

9. There is such a thing as an arrogant, conceited, idiot.

10. Bugs happen.

11. Any technology which doesn't make it in the first year, won't.

12. Rule for technological success:
For a product to beat out an encumbent product in the same space, it must not violate any of the following rules:

1) It must be at least 3 times better than it's competition
2) It must be easily understandable
3) It must be affordable

Violation of any one of these rules will spell disaster for a new technology no matter how cool.

13. People tend to immediately blame that which they do not understand, usually due to their own ignorance.

14. Most people will continue to use that which they are used to despite the presence of a superiour product.

15. Rules you must know in order to appear as a genius to your co-workers:

1) Never admit that you don't know something, there's always plenty of time to look it up.
2) Never, under any circumstances, document anything. This impacts job security.
3) Always hang out with the other so called "geniuses" on the project, so that you,
by association, will be considered a genius.
4) Always eschew the opinions of others as nothing, meaningless and beneath notice.

16. People will use your software in the most obtuse, indirect and unexpected way imaginable.

17. "Why specifications survive" -- Inspired by something I once read, but can no longer find... Ever wondered why bad specifications survive? Here's a good example. Take a look at the standard railroad track. It's about three feet across. Well, the reason it's about three feet across is because the people who were employed to make the railroads were wagon makers. These wagon makers were usually immigrants from different parts of Europe who, of course, made wagons there. Well, the wagons in Europe needed to be a certain width, to fit the ruts in the road which exist due to hundreds of years of wear and tear. These ruts were caused by Roman chariots from when significant portions of Europe were under the control of the Roman Empire. As it turns out, a chariot is about as wide as the hind quarters of a horse. So the next time you wonder what horse's ass came up this this design, you'll have something to reflect on.

18. Untested code never works the way you think it will, so please test.

Tuesday, June 14, 2005

Love/Hate with IBM

Okay... ever since I heard that IBM is more or less responsible for the current Apple PowerPC->Intel switch, I'm a little pissed off at IBM. IBM doesn't see any profit in making thier processor faster for Apple unless Apple invests heavily in the PowerPC's development. That is to say that Apple needs to pay handsomely AND IBM will own the result, that wreaks.

It seems, from all accounts, that Apple was pretty much painted into a corner. They couldn't make the profits they wanted without the 3GHz PowerPC desktops and G5 laptops, but they would have to give up that profit to IBM. That sucks.

Okay.. so now I hate IBM because of Apple, but I love IBM because they are defending the OS community against people like SCO. My heart is torn to pieces.

Saturday, June 11, 2005

Comcast calls right on que

Sometimes you have to wonder. They called me on Thursday, right after Verizon fixed their problem. I really think they must have known or something.

Anyway, the guy is supposed to come today to set things up, so I'm here waiting for him.

My Disappointment with Apple

Although Apple moving to Intel is good news, it's terrible news that Mac OS X will not run on plain vanilla Intel machines. Steve and Co. really need to get their act together and let this fly.

If people could buy OS X for their PC I bet you would get a lot of people switching to OS X who would never have thought about getting a Mac before. If the "essence of the Mac" is truly the operating system, as Steve himself said in the keynote speech, then why should Apple be afraid of this prospect?

Wednesday, June 08, 2005

Apple goes Intel -- Implications and predictions

Wow. In a stunning move Apple has decided to go to Intel processors. Being an ex-NeXT guy I can tell you that I've always known that OS X was leading a double life on intel processors. This might actually be some very exciting news.

A few questions come out of this event:

Q1) What does this mean for Linux?
Q2) How will this effect Windows?
Q3) What does this mean for GNUstep?

A1: Basically nothing. Mac OS X cannot compete with the price of Linux or the *BSDs. It's $129. FLOSS will always be very cheap or freely available.
A2: It will likely have a good deal of impact on Windows. There is already a project called MacWINE whose purpose is to allow the Mac to run Windows programs. Once OS X is running on Intel, I predict that this project or something very much like it will become a centerpiece of OS X's collection of standard programs. This would remove the common excuse most people give for not using a Mac: "There's no software for it."
A3: This could be an opportunity for GNUstep to shine. If GNUstep can become the equivalent of Wine for OSX, then it bodes well for GNUstep. Additionally, Apple is unlikely to port its APIs elsewhere (i.e. to other environments) so GNUstep will be necessary to fill this gap.

Busy week

Verizon really needs to care more about it's customers. My DSL went down on Sunday morning at 1:30AM and stayed down until yesterday. It has been intermittent since then. While it seems solid now, the experience has shaken my confidence on Verizon greatly.

I'm wondering if it might not be time to switch to a cable modem.

Foundation is now close to Catalina compatibility

I have worked hard to get it to this point, but all of the classes in Catalina are now present in GNUstep's base implementation. Soon, ...