Special collaboration by Camilo Telles; article originally published in tellEsfera.
I'm the kind of guy who loves to show the good things that happen in Brazil, the stories of engineers and technicians who produce or have produced things in our country.
Taking advantage of this motto, I decided to recall a little-known history of national computing: the Brazilian Macintosh (called Mac 512). I decided to do this by interviewing Rainer Brockerhoff, a hacker with a full hand who knows the Mac Ministries as I believe no other Brazilian knows and who participated in the development of the Brazilian Macintosh.
I did this in the 10-question model, which you can see below:
1) What is your history and how did you get into the Brazilian Macintosh project?
I started working as a programmer in 1969. At that time, there were practically only computers in banks and universities. In 1977, I bought an Apple II for home use and, a few years later, when the first national microcomputers started to appear, I started working at one of the manufacturers, Quartzil. There, I was responsible for the operating system and also learned to design hardware.
When the Macintosh was launched in 1984, I became interested and brought one of the first to Brazil. I also brought development tools and started making small programs for my own use. I already knew Unitron people from the Apple II era. When the Mac Unitron hardware was almost ready, I was invited to help with the software, and of course I was interested.
2) How was the team composed and what was your role in the team?
I think there were about 10 people. I was an external consultant, as I live in So Paulo, and went there once a week. I made a good part of the Toolbox the part of the ROM responsible for interacting with applications and with the user. Other people did the device drivers and the graphics routines. I must have done maybe 30% of the ROM; It is difficult to evaluate today. Later on, I was also responsible for the boot ROM, the equivalent, at the time, of the firmware and I was also invited by SEI (Special Informatics Secretariat) to prepare a technical report detailing how the Mac Unitron software had been engineered.
3) How was the software reverse engineering process carried out?
A California programmer, Steve Jasik, had developed a product called MacNosy (proboscis) to decode the Mac's ROM. We used that a lot, but there were also other tools, the exact name of which I don't remember anymore.
4) How was the development process? What tools were available? Which platform did you develop on?
It was a process of making call by system call. For each one, I took that part of the MacNosy output, which was in Assembler language, not very simplified, and made notes or changes to make the logic more intelligible. Include in this listing patches (changes) introduced by the system diskette to fix bugs or expand functions, and compared that to the description of that call in Apple's documentation. So I recoded that function in C language, we had a compiler called Aztec C that was very reasonable. Then, it checked if the object-generated code really performed the desired functions.
It all ran on my Mac 512K. At a certain point, we had a ROM that could already be tested. This was possible, too, because the Mac Unitron had twice the available ROM space of the Apple Mac. Apple programmers had to use many tricks to make the software fit, while we had space to absorb the inefficiencies of C and still fix several bugs right in the ROM.
5) You lived through the market reserve period. In your opinion, what were the pros and cons of that time?
I think it was a wrong and inadequate reservation, which did not reach its objectives; especially since few people, at the time, understood the technical aspects or predicted the progress of globalization. All were based on industries that took decades to establish and did not foresee the acceleration of digital technology. It was not feasible to manufacture chips in Brazil, but it could not be imported that delayed the deployment of cars with injected engines for more than a decade, for example. In the company where I worked, ostensibly protected by the reserve, we needed a logical analyzer to develop the system. The analyzer contained a microprocessor, so it could not be imported without an exemption process that took almost three years! As the entire industry was in this situation, the reserve was a major contributor to smuggling.
6) The Macintosh project never came out due to government interference. Can you tell us a little about this story?
As I said, I made a technical opinion detailing that the project was legal within the concepts, at the time, of reverse engineering. I followed the rest only through third party information, but what I was told was that SEI made two favorable technical reports examining hardware and software separately.
During the approval process, under American pressure, Congress passed Law 7646 (Software Law), which further delayed things, and the project had to be redone and resubmitted. In 1989, CONIN rejected the project. CONIN was composed of eight representatives of civil society and eight government ministers. Seven independent representatives were present and voted in favor of the project. Seven ministers voted against and one abstained. In view of the tie, the Minister of Science and Technology, Chairman of the Commission, voted against.
7) What is your opinion about the legal limits of reverse engineering?
I'm not a lawyer, so that's just an opinion. In the Unitron case, the Brazilian market was closed to Apple, and she had not registered patents here. And obviously Unitron would not be able to sell on the North American market; so it was a dispute over intellectual property concepts. Reverse engineering was done with full access to the original, which today would not be accepted; but within the concept of reserve of the time it was valid.
8) And after the Macintosh project? What did you do?
For several years, I was the technical director of a company that manufactured digital medical monitors, which was even an example of how technology could be developed here without copying anyone and competing with imported devices. In parallel, I set up one of the first commercial internet providers in Brazil. Of course, I have always consulted and developed software for the Macintosh. Today I consider myself semi-retired, but I still do shareware for Mac.
9) You are one of the rare Brazilian developers for Macintosh. How do you develop for this platform and how do you get your products to market?
I think the most important thing is to target the global market, work only on the internet and stay in touch with the developer community. It is necessary to master English very well, of course. I chose a niche market that facilitates this shareware for users of better technical level and for fellow developers. I often go to congresses abroad and publish several free software and / or open source. All of this generates publicity and recognition by the community. None of this works if the products are not well finished and functional. The Mac market is very demanding in that sense. My family tradition of artisanal joinery: my father, for example, specialized in producing wooden models for foundry that had to be extremely accurate. I am the first non-joiner in the family, but I inherited the obsession with polishing and perfecting my products to the maximum. If I needed shareware to survive and devoted full time to it, it would certainly be possible especially now, with the Apple market exploding in several directions.
10) Apple has been substantially increasing its presence in the market. You went through the ups and downs of the company. How do you think this new moment will change the lives of developers? Do you already notice differences in the adoption of your products?
Of course, the market tends to increase, and that's a good thing. For developers, resources are always improving; I now only make products for Mac OS X 10.5 onwards. With the iPhone / iPod line, a huge number of developers from other platforms have arrived, some very beginners. I think that the most adaptable of these will later start making software for Mac as well, and it is always good to grow the community. An advantage we have here is that this does not necessarily mean greater competition over a few niches, but rather great opportunities for collaboration in exploring new niches.
. . .
Some technical information, via OLD-COMPUTERS.COM Museum and corrected / supplemented by Rainer:
|Keyboard||Complete QWERTY with 58 keys|
|Graphics modes||512 × 342 points|
|Sound||Same as original Mac 512 (1 channel)|
|Dimensions / weight||Much like the original Mac 512|
|Built-in media||3.5 inch floppy drive (double sided, 800K)|
|Operational system||System (Mac OS translated and adapted by Unitron)|
(Some photos in this article are by Carlos Duarte do Nascimento, better known by his nickname, “Chester”. The machine in question was still a prototype of Unitron.)