ADVERTISEMENT

Dialog Box

By Matt Neuburg, Managing Editor

Bring Back the Good Old Days

I am an entirely self-taught, by now intermediate level Mac programmer (unabashedly Pascal - CW, though I still utilize Think Pascal 4.0.2) who enjoys MacTech very much. I have been reviewing and learning a great deal from your back issues from the “distant” past! Nevertheless, I do have certain beefs. I am somewhat disturbed by the trend to feature to such a great extent the more technically advanced aspects of programming understood by the few, from which the rest of us can learn very little, I’m afraid. In the past there was a certain gleeful joy about programming, a certain spirit and amazement and perhaps naiveté about it all that seems to be missing now. The magazine seems too severely oriented to the well-equipped professional, with an insufficient amount of time devoted to us weekend, sparetime adventurers.

Additionally, far too much space is given over to Symantec and not enough to Metrowerks. And there is just too %^&* much C++. As a “Pascallian”, I read a lot of C and adapt it to my needs. C++ is impossible, and I find it very hard even to follow the basic ideas being presented. I know that you cannot force people to write articles in any one language (and I know that C++ and C for that matter aren’t going to disappear), but it would be great if you encouraged authors whose articles are in C/C++ to make some effort to put the Pascal/OP counterpart in a parenthetical or a footnote. At least some effort ought to be made to explain the less obvious portions of the C/C++ program code by pointing out various things that Pascal programmers should be aware of and by offering possible Pascal/OP solutions where that is possible. After all, as was noted in the November issue somewhere, MacTech is the only programming magazine for the Mac! Efforts should be made to include more of us in the challenge that is Mac programming. I, for one, don’t like feeling left out for no good reason. Thanks.

- Jeff Niederhoffer

[Jeff, I sympathize strongly with your plea. At MacTech, we’re looking to extend our scope to include once more those “weekend, sparetime adventurers” - without, of course, losing our edge as a forum for the professional developer community. Of course, sending us the articles is up to you, the readers! - Ed. man]

[As for your comments on Symantec vs. Metrowerks, please take note (in this issue) of the introduction of “From the Factory Floor” - our new column to give Metrowerks a presence in the magazine. You should expect to see this as a regular feature and you should take pride in knowing that you, the readers, made it happen. - Ed. nst]

Apple Missing its Chance to Rewrite the OS

Scott Boyd hit the nail on the head with his comments, in the November issue of MacTech, about overcoming limitations. Apple has made a tragic mistake in the way it is handling compatibility with Copland. The legacy code of Apple continues to grow, with all its limitations. It is very sad that developers are constantly faced with constraints that could have been built out of the system years ago.

Apple needs to look at the OS and see that when a machine has 8-16 meg as standard, 32K worth of text (or less if it’s a big font) is ridiculously small. Apple has also recommended that we avoid creating too many resources (I can’t recall the exact figure, but I believe it is in the 2K range). In Apple’s words, “The Resource Manager is not a Database”. My question is: Why not? Why can’t we have a 10,000 entry file that contains 60K of text that can be written in a few lines of code? Why do we need to do linear searches to find a resource?

How could Apple have fixed these problems? One obvious way might be to create new Managers with new APIs. I’m sure developers would flock to such managers. While this would work, I think Apple could have done something more radical to rid itself of all this legacy code.

Apple could have created Gershwin from the start, a real multi-tasking OS. For compatibility, it could create a virtual System 7 machine, much like MS Windows creates a virtual machine for DOS programs. This could have been a single thread in Gershwin that emulated a System 7 environment, with 68k emulation and all. Gershwin programs would be 100% native, running on a pre-emptive native OS. If you need to run System 7 software, you have that too. The toolbox APIs could be completely rewritten, no constraints required. Likewise, Apple could have written a new file system that fixes many of the limitations of HFS. The File System Translation Manager would support old “inferior” file systems, much in the same way as we can read and write DOS disks.

Of course, we know none of this will happen. And unfortunately, it also means that we won’t see a truly pre-emptive MacOS for at least 4 to 5 years. Boy, that Mac will be great in the year 2000; I can’t wait. Don’t get me wrong, I love my Mac. It is just disappointing that Apple has stumbled so much for so long.

- Mark Munz, Puppy Dog Software