Latest press releases

Plextor announces its first Serial ATA 12x DVD writer

Intel Second-Quarter Revenue $8.05 Billion

Intel Xeon Processor Family Ushers in New Technologies, Broader Range of Capabilities for Enterprise

The First Xeon EM64T (Nocona) Workstation/Server Solutions


Interview with Fleecy Moss, CTO at Amiga Inc

Author: Vidar Langberget.
Posted : 2002-05-03 00:00:00.0
Category :

SDK, x86, IDE

3) There's a rumour going around that Amiga don't have access to the source code for OS3.9. Is that correct?

There are rumours that Elvis works in a Wal Mart in Wyoming, that the moon is made of cheese and that Tony Blair wears a wig. Rumours are what they say they are.

4) When the OS4.x project was announced, OS5 was said to be ready late 2002. Is this still the target date for OS5? If not, when? Has development started?

Given the burn we all got over the 01 November situation, I don't think it would be responsible anymore to say anything beyond we are committed to the future of the AmigaOS.

5) OS5 is said to offer VM, multiprocessing, memory protection and other advanced features. Will OS5 be developed from OS4 and AmigaDE, but without regard to backwards compatibility, or will it be written from scratch?

It will provide a solution that doesn't yet exist in the world for a problem that has never really been solved, namely how to make technology invisible.

6) The release of Amiga XL and Amithlon combined with the delays of the PPC machines have had a few people suggesting that an x86 version of Os4.x should be reconsidered, and that Amithlon could form the basis for this development. Has this been considered lately? There are ways to stop people from installing other OS' on Amiga x86 machines, something I know that developers like Hyperion are concerned about. And "custom" x86 machines(using standard motherboards etc, but with a special BIOS, or something similar) would make it less risky to release new Amigas for Amiga hardware companies and it would be easier to keep up with developments in hardware. Any thoughts?

Hardware is a commodity. We have that attitude with the AmigaDE and we will have that attitude with AmigaOS5. AmigaOS4 will be the OS on the AmigaOne.

7) A new release of the SDK, with the first real pieces of Amiga developed code has been promised for a VERY long time. When can we expect this update?

SDA developers have had access to lots of the pieces for quite some time now and three of them have even been playing with the first pieces of the Amiverse - the service broker, the content service, the description service, the activity service. The DE player itself has other pieces in it, and the next release of the DE player will have many other pieces. Zeoneo has already written dynamically scaling versions of Convex and PlanetZed - if you have enough hands, or two people, you can actually play PlanetZed and be scaling it whilst dodging aliens. Pulsar Interactive's Gobbler used the Ami2D animation services and that was released months ago.

The issue isn't where is the stuff - any developer serious about developing the DE would already have the latest builds and have access to all of these pieces. The issue is bringing that all together along with proper documentation, a wad of test and development tools etc etc. I could burn a CD now with all our stuff on it but an SDK needs to be much more.

So as to when - it will be out in the future.

8) The GUI on the DE has been criticised a lot. When will we see a new GUI? Serious applications for the DE seem to be lacking because developers don't want to waste time writing applications using a GUI that will be replaced soon.

What DE? No one about from a few OEMs and developers have even seen the DE. If you mean the SDK, then that's the Tao-Group's toolkit. We have our own toolkit as well as Prism, a descriptive interface engine, which again, has been available since theee beginning of the year. In addition, we are developing our own interface system which will simplify, enhance and wrap the Tao toolkit. The problem we have with replacing it is that the J-Engine is tightly tied to it and I don't think too many people want to throw away Java capabilities.

9) Most modern AmigaOS programs use MUI. Have you talked to the author about a PPC version for OS4.x?

AmigaOS4.x uses a greatly enhanced Reaction.

10) The DE player and a number of games have been available for quite a while now. How has sales been?

The US editor of FamilyPC sent us a bug report for Boxicon, we get quite a few game downloads from the Microsoft campus, so we are happy considering we haven't really started to market the direct sale PC element yet. We are concentrating on the STB and PDA companies, with bundling opportunites for millions of units.

11) It was announced a few months back that Sharp would use the intent JavaVM and the DE for their Linux PDA. Now, it seems that Sharp have selected another JavaVM for their PDA. What's going on?

You'll have to talk to Sharp and/or Tao about that. I will have my Zaurus at Alt-Woa and it plays the DE content just fine.

12) What makes ACM better than CORBA and other similar technologies?

The ACM isn't CORBA, which is a service architecture - it is a component model, more like COM. It is used like lego to build systems by plugging bits together. It is better than COM because it is fully introspective and identifiable, rather than relying on internal mechanisms like iQuery, and because it doesn't have all the nasty hooks and baggage that comes with trying to support legacy solutions.

It supports pluggable resource management, allows for thread safe and non thread safe development and has a completely clean separation between description and implementation. What is holding us up at the moment is the documentation more than anything else, since it requires a lot more than a few cursory code examples. It is a very different way of doing things.

Next : Test 2