Previous Issue | Search TidBITS | TidBITS Home Page | Next Issue
Copyright 1990 TidBITS Electronic Publishing. All rights reserved.
Information: <firstname.lastname@example.org> Comments: <email@example.com>
Two connotations of the above title come to mind. ROM liberation and ROM libraries. Both are apt, because ROMlib is a Unix library that can simulate the Macintosh ROMs, but which might liberate (perhaps not the best choice of words, I suppose) them from Apple. For those who aren't up on any of the latest clone reports, this is an extremely interesting (and potentially litigious) product, because Macintosh source code can be re-compiled on a Unix machine. The ROMlib version of the program retains the complete look and feel (I hear legal knives being sharpened already) via X Windows. The advantages of doing this are the advantages inherent to Unix machines - protected virtual memory, Unix development tools, "true" multitasking, the Unix file system, and heavy-duty hardware. Of course the main advantage is that there are a huge number of useful, elegant Macintosh applications that can be useful and elegant on a Unix machine.
Needless to say, there are a few small problems. Compatibility with the Mac Toolbox calls only extends to volumes 1 through 4 of Inside Mac, though the company, Abacus Research & Development Incorporated (ARDI), is working on volume 5 and 6. Other stuff that isn't supported includes the Printing Manager, Desk Manager, Device Manager, Disk Driver, Sound Driver, Serial Driver, AppleTalk, Disk Initialization and SCSI Manager. The package itself has been ported to a number of Unix boxes, but is only supported on the Sun/3s because of ARDI's current size of four programmers, a lawyer, an accountant, and a secretary.
ARDI does expect to be sued by Apple because they are aiming to sell ROMs for Macintosh clones, but they feel that they are completely in the right. They think the issue boils down to whether or not Apple owns the look and feel of all Macintosh applications regardless of who developed the programs. It seems to us that a more dangerous possibility is that Apple can claim copyright on the specific names of the routines and thus prevent ARDI from using those names. Since ARDI has completely reverse-engineered those routines, they are safe on the source code copyright aspect, but not being able to use the same routine names would be a pain because it would force programmers to change all the routine names before compiling under ROMlib (though I suppose ARDI could just release an application to change all the routine names appropriately).
ROMlib costs $400 (Oh, a round number price!) for a single-user license and is currently being bundled with Executor, a non-product that can run a good number of Mac binaries on the Sun/3. ARDI wrote Executor in the last four or five months to test ROMlib. Now that they have released ROMlib, they are devoting more attention to Executor. They even posted a list of applications from the BCS and BMUG CD-ROMs that they have tested under Executor and a number of the programs ran relatively well. None of the major commercial programs that they tested (Word, Excel, Wingz, SuperPaint 2.0, HyperCard 1.2.2, or MacWrite) ran without a hitch, but all but Wingz often break whenever Apple does anything different in hardware or software.
Abacus Research and Development
1650 University Blvd.
Albuquerque, NM 87102
Clifford T. Matthews -- ARDI
Walt Leipold -- firstname.lastname@example.org
Adam C. Engst -- TidBITS Editor
It's not exactly the San Andreas Fault, but IBM and Microsoft have been getting along poorly, with the latest spat concerning IBM's new version of OS/2 that runs in 2 megs of RAM. The problems began the fall of 1989 when Microsoft pressured IBM to drop OS/2 "Lite," aimed at DOS users. In April of 1990, IBM said that Microsoft would be taking over most of the OS/2 development work. However, when Windows 3.0 came out in May, essentially destroying the market for OS/2 at many sites, IBM was not among the PC makers who announced plans to bundle Windows with their hardware. In June, IBM started showing off its own 2 megabyte version of OS/2, which was snubbed by Microsoft.
OS/2 battles are not the only place where IBM and Microsoft have been at odds in the past few years. During the hubbub with Apple and Microsoft announcing TrueType for Macs and PC-clones running Windows, the ever-conservative IBM endorsed PostScript for OS/2 to Microsoft's dismay. IBM's choice of PostScript was interesting especially in light of a less conservative decision to license Steve Jobs' NeXTStep environment, which use Display PostScript for on-screen font rendering.
A final battleground is shaping up on the handwriting technology front, because Microsoft is attempting to develop its own handwriting recognition technology (no doubt to be integrated with Windows some time in the future), while IBM chose to go outside again, licensing Go Corp.'s technology. Such unfriendliness.
Unfortunately for those of us who aren't specifically tied to IBM or Microsoft products in any way, the two companies still push the market around. The good aspect of the growing discord is that IBM and Microsoft together may well be unstoppable; separated, they carry far less clout. In some ways, it was too bad the Lotus/Novell deal fell through, because they (along with WordPerfect) were the only companies really strong enough to force Microsoft to play nice and not try to take over the entire microcomputer industry. Such a takeover must not be allowed to happen because even if a Microsoft monopoly was benign in nature (if not in interface), it would certainly squash a good deal of innovation from smaller developers who would be unable to compete. These days, a great product does not guarantee a share of the market, especially when it's a market from which Microsoft profits.
Adam C. Engst -- TidBITS Editor
InfoWorld -- 27-Aug-90, Vol. 12, #35, pg. 1
PC WEEK -- 30-Jul-90, Vol. 7, #29 , pg. 1
Since TidBITS is distributed only electronically (at least by us, others may re-distribute in other ways), many of you have probably come to rely on electronic mail. Most email runs on mainframes or workstations, not because they are better suited to the task, but because they are more often connected to networks. Until ISDN become available most places in the world (for those of you in Europe who already have it, don't hurt yourselves chortling :-)), the rest of us will have to rely on other methods of connecting our Macs to the e-world.
Until recently, the process of setting up email on a Mac was relatively expensive with packages such as CE Software's QuickMail and its accompanying bridges. The best alternatives were arcane ports of Unix's uucp software with minimal Mac interface options. Since MacTCP has been out though, more mail packages are showing up because MacTCP makes it much easier for a Mac to communicate with a host computer also running TCP/IP. The two latest programs, Eudora from Steve Dorner and MacPost from Lund University in Sweden, share the admirable feature of being free.
Eudora is at version 1.1 and is available via anonymous FTP from ux1.cso.uiuc.edu in the mac/eudora subdirectory. Eudora can use the POP3 (can't for the life of me remember what this stands for, Post Office Protocol, perhaps?) and SMTP (Simple Mail Transfer Protocol) protocols. More importantly for those of us not directly linked to the Internet, Eudora works with either MacTCP or the Communications Toolbox, though Steve Dorner says "there are some gotchas in the latter." Feature include nicknames, multiple mailboxes, automatic mail checking, and automatic binhex/de-binhex of Mac documents. Source code and a 60 page manual are both available, and all Steve asks is that you tell him that you are using Eudora and that you send him any feedback you may have. I haven't gotten a chance to try Eudora personally, not having access to either the Comm Toolbox or MacTCP, but John Norstad, author of Disinfectant, said he uses Eudora and finds it extremely useful and stable.
The other package, MacPost 1.0b2 is the result of a joint project between Apple and Lund University. It is a bit heavier on the hardware side because it runs a client/server which requires a dedicated Mac running MacTCP as the server. The server must also be connected to an SMTP server on the Internet. MacPost is available via anonymous ftp from pollux.lu.se (126.96.36.199) in pub/mac/comm/macpost. The client and server communicate over AppleTalk, which limits the extent of the server to an AppleTalk internet, but dial-in access through Liaison or Shiva's NetModem is possible as well. So if you are looking for email access to the Internet and can get a connection, check out these package. If nothing else, they're more than worth the price you pay.
Steve Dorner -- email@example.com
John Norstad -- firstname.lastname@example.org
MacPost Mailing List -- email@example.com
Before anything else, I want to mention that the most recent issue (September) of BYTE magazine is mostly devoted to the discussions of many of the notable figures in the computer and electronics industries. The subject is the past, present, and future, with an emphasis on the future. If you are at all interested in these industries as more than a casual user of their products, I highly recommend that you check out this issue.
Here's a bit from Michael Kessler on the HUMANIST discussion list that shows the perils of computer use. Nasty colloquial Find/Replace feature there... :-)
"A correction from the pages of the Fresno Bee: "An item in Thursday's Nation Digest about the Massachusetts budget crisis made reference to new taxes that will help put Massachusetts 'back in the African-American.' The item should have said 'back in the black.'""
A few things have changed in this issue of TidBITS, most notably the display font. We used to use Bookman 12 point, because it is a good, readable font. However, it is a tad large, which results in not much text being displayed on the screen at once. From now on, we'll be using New York 10 point instead, although you are perfectly welcome to change it for each issue you get. We've known about this problem for some time now and had planned to fix it relatively quickly in TidBITS II. Unfortunately TidBITS II will use some of the new features advertised for HyperCard 2.0, which is obviously why you haven't seen TidBITS II. We've also been very busy with other projects. So changing the font is a bit like sitting down to read a good book before people are arriving to visit you - they will come all the sooner so you can't read much. Perhaps by changing the font, HyperCard 2.0 will appear immediately to allow us to start working on TidBITS II. And if HyperCard's release date continues slip, we offer the top ten excuses direct from the HyperCard 2.0 development team.
Speaking of TidBITS II, there are some interesting ideas we're playing with. Potential enhancements include the ability to import and export subtly-tagged text files so the distribution files will be human-readable. We'll also try to build in filters so popular types of online text, such as the Info-Mac digests and clippings from Usenet, can easily be archived for future reference. To do this, we hope to utilize techniques in HyperCard that should increase performance greatly in large archives. The idea is to turn TidBITS into a general-purpose text archiver and reader so we as an electronic community can store and retrieve our textual information quickly and easily. Just think, wouldn't it be nice to have all those "readme" files in one place, easily located and searched even if you don't have the right word processor handy?
We always appreciate comments and always respond if electronically possible. Now even more than before, we would like suggestions for future enhancements. We've saved all the suggestions you've sent already sent us, so those comments will be taken into account. So please send mail with your thoughts.
Adam C. Engst -- TidBITS Editor
Michael Kessler -- Michael_Kessler.Hum@mailgate.sfsu.edu
BYTE -- Sep-90
The Top 10 Reasons HyperCard 2.0 Has Not Yet Shipped
Disclaimer: It's all lies. Lies lies lies.
These lies are fictitious. Any similarity to actual lies, fibs, or prevarications is purely coincidental.
These lies are the property of the HyperCard Development Team. Any rebroadcast, retransmission, or use of the pictures, descriptions, and accounts of these lies without the express written consent of the HyperCard Development Team would really be a bad thing, and in poor taste too.
Additional Disclaimer: These are last month's lies. This month's lies are very different.
10. Bill left without telling us how it worked.
9. We were saving it as a going-away present for Jean-Louis.
8. It took months to get the color out after we discovered that the manuals didn't mention it.
7. We introduced it at the annual developers' conference, and we thought there was a rule that says that anything you introduce at a developers' conference you can't ship until after the next developers' conference.
6. Bowling shirts just take longer than T-shirts.
5. For most of us, it was a great way to avoid sweltering in Boston in August.
4. It took months to devise all those phony seed releases, with all those phony bugs, which we were doing only as a clever ruse, of course.
3. Nobody told us you were supposed to finish the thing first.
2. Howard Spira had his money on System 7.0.
1. We couldn't ship until we had tested it with the new Macintosh LX, the one with the impressive performance, quiet ride, and distinctive styling, all for under $25,000. See your dealer today.
Kevin Calhoun -- firstname.lastname@example.org
Non-profit, non-commercial publications and Web sites may reprint or link to articles if full credit is given. Others please contact us. We do not guarantee accuracy of articles. Caveat lector. Publication, product, and company names may be registered trademarks of their companies. TidBITS ISSN 1090-7017.
Previous Issue | Search TidBITS | TidBITS Home Page | Next Issue