home *** CD-ROM | disk | FTP | other *** search
I've heard many times how ARDI designed Executor from the ground up, using clean room techniques, because of the proprietory and secret nature of the Macintosh toolbox ROMs. Hence programs which use certain undocumented features of the ROMs will not work on Executor. How then, do you explain the new Power Mac clones, from companies like Radius? These machines are supposed to be 100% Mac compatible. Do they reverse engineer the ROMs, do they buy the ROMs from Apple, or has Apple made avaiable the code for these ROMs in the "Open Reference Platform" or something like that. I'd be interested to know what Radius and the other clone maker(s) do in this regard. Bye.... -- ________________________________ John de Bruin Email: j.bruin@auckland.ac.nz From owner-paper Fri Jan 5 03:14:18 1996 Return-Path: <owner-paper> Received: by ftp.ardi.com (Smail3.1.29.1 #3) id m0tY99t-0007qCa; Fri, 5 Jan 96 03:14 MST Sender: owner-paper Received: from ardi.com by ftp.ardi.com (Smail3.1.29.1 #3) id m0tY99C-0007q3n; Fri, 5 Jan 96 03:13 MST Path: sloth.swcp.com!usenet From: Clifford T. Matthews <ctm@ardi.com> Newsgroups: comp.emulators.mac.executor Subject: Re: Radius Mac clones Date: 05 Jan 1996 03:09:40 -0700 Organization: ARDI Lines: 79 Message-ID: <ufhgybrlzv.fsf@ftp.ardi.com> References: <30EDE214.5C26@auckland.ac.nz> NNTP-Posting-Host: ftp.ardi.com In-reply-to: John de Bruin's message of Fri, 05 Jan 1996 18:44:36 -0800 X-Newsreader: Gnus v5.0 To: executor@ardi.com X-MailNews-Gateway: From newsgroup comp.emulators.mac.executor Sender: owner-paper@ardi.com Precedence: bulk >>>>> "John" == John de Bruin <j.bruin@auckland.ac.nz> writes: In article <30EDE214.5C26@auckland.ac.nz> John de Bruin <j.bruin@auckland.ac.nz> writes: John> I've heard many times how ARDI designed Executor from the John> ground up, using clean room techniques, because of the John> proprietory and secret nature of the Macintosh toolbox John> ROMs. Hence programs which use certain undocumented features John> of the ROMs will not work on Executor. I'm sure it sounds like an old song and dance to people who have followed us closely, but it's the truth and it explains questions that come up frequently, so you'll probably hear it more, too. Technically we don't really need the ROMs, per-se, we only need a functional specification to work from. In other words as long as we know what every single routine has to do, and all the side-effects that applications could count on, we could fairly easily code up equivalents. The way "clean room/dirty room" engineering works is you get "dirty" engineers to examine ROMs and write functional specs and they then pass those specs to a bunch of lawyers who make sure that the specs are functional in nature and do not reveal any implementation details. The lawyers then give the specs to the clean room engineers who then implement according to specs. This is expensive, but it's doable. It would take any company other than ARDI too long to do this though, due to the amount of work required. Luckily, we wouldn't really have to implement much new stuff, just make sure all the side-effects were properly handled. John> How then, do you explain the new Power Mac clones, from John> companies like Radius? These machines are supposed to be John> 100% Mac compatible. Do they reverse engineer the ROMs, do John> they buy the ROMs from Apple, or has Apple made avaiable the John> code for these ROMs in the "Open Reference Platform" or John> something like that. They have not reverse engineered the ROMs. There have only been a very small handful of companies who have tried anything even vaguely similar to what we've done, the most notable were Nutek and Quorum. Both had much less compatibility that we have and went (virtually) bust trying to do what we have done. Radius and PowerComputing use Apple's ROMs via a license from Apple. We approached Apple's licensing department and were given the names of a couple people to talk to, but they did not return our e-mail. During MACWORLD Expo you can bet that we'll be talking (at least unofficially) to many Apple employees. Apple's official position is that they will not license any 68k based stuff, so we'll have to get a PPC emulator going before we can officially license Apples's software (unless they change their mind). Luckily, VCPU will make a PPC emulator relatively easy, *and* even without Apple's permission we can make Executor so that you will be able to drop a copy of System 7.5 on top of it and have it work. That requires a lot of work -- more than the few engineers we currently have, *but* it's work that we understand and we have a tremendous framework to start with, so it's largely a matter of getting money for engineers, which is something else we're pursuing at MACWORLD Expo. John> I'd be interested to know what Radius and the other clone John> maker(s) do in this regard. My understanding is that they not only license the ROMs, but they also license much more, too, including ASICs that are in such short supply that Gateway wasn't able to cut a licensing deal with Apple because they weren't able to turn out the ASICs quick enough. John> Bye.... -- ________________________________ John> John de Bruin Email: j.bruin@auckland.ac.nz I hope this explains some of the legal issues involved. These are all matters that we're pursuing with vigor. --Cliff ctm@ardi.com From owner-paper Fri Jan 5 04:14:25 1996 Return-Path: <owner-paper> Received: by ftp.ardi.com (Smail3.1.29.1 #3) id m0tYA64-0007qCa; Fri, 5 Jan 96 04:14 MST Sender: owner-paper Received: from ardi.com by ftp.ardi.com (Smail3.1.29.1 #3) id m0tYA5G-0007q3n; Fri, 5 Jan 96 04:13 MST Path: sloth.swcp.com!usenet From: Clifford T. Matthews <ctm@ardi.com> Newsgroups: comp.emulators.mac.executor Subject: Re: Where to send change of address... Date: 05 Jan 1996 03:38:12 -0700 Organization: ARDI Lines: 63 Message-ID: <ufg2dvrkoc.fsf@ftp.ardi.com> References: <4c61jv$dp6@turing.mathworks.com> NNTP-Posting-Host: ftp.ardi.com In-reply-to: peter@mathworks.com's message of 31 Dec 1995 13:01:51 GMT X-Newsreader: Gnus v5.0 To: executor@ardi.com X-MailNews-Gateway: From newsgroup comp.emulators.mac.executor Sender: owner-paper@ardi.com Precedence: bulk >>>>> "Peter" == Peter Greis <peter@mathworks.com> writes: In article <4c61jv$dp6@turing.mathworks.com> peter@mathworks.com (Peter Greis) writes: Peter> I guess that this one is for Cliff... In general, counting on Cliff, or any single ARDI employee, is a bad idea. It is not uncommon for me to get two or three weeks behind in my e-mail or Netnews reading when I am working on getting a new release out (as I was getting 1.99p8, 1.99p9 then 1.99q out). As such, the generic addreses are much better, since more people have a chance of being able to respond to them. After all, when I am away, nobody reads my personal e-mail. For the record, here are the major generic addresses: bugs@ardi.com -- Sending e-mail to this address is preferable to sending a bug report to c.e.m.e. or the Executor Interest mailing list, although the absolutely best way to send a bug report is via http://www.ardi.com info@ardi.com -- This will eventually be an automated information server. Right now it's equivalent to questions@ardi.com questions@ardi.com -- ARDI's tech. support. Currently we tend to answer all questions, although since we're incredibly overworked, once Executor 2 is in beta, we'll start answering only questions asked by registered users, although we'll try to keep the FAQ current to help everyone out. sales@ardi.com -- For specific questions regarding pricing, availability, site licenses, dealer pricing, etc. Since there's no address specifically for changing addresses, sending an address change to questions@ardi.com is the best bet. NOTE: I've cc'd this to questions, so someone will indeed see it and get your new information (if nobody has done this already). As ARDI grows we'll have to compartmentalize. Right now tech. support, web-maintenance and sales are both handled by Melissa, but if you use the e-mail addresses above, the right thing will automatically happen when we hire more people. Peter> who do I send a Peter> registration change of address to? I have a license for one Peter> of the early versions of Executor (NeXTSTEP, of course), Peter> now at version 1.3b. I know that somehow I need to get a Peter> new registration string, the old one does not work. Peter> Keep up the great work! Thanks. '95 was a very good year for Executor's development. In '96 we hope to do much more. Peter> -peter --Cliff ctm@ardi.com From owner-paper Fri Jan 5 04:14:27 1996 Return-Path: <owner-paper> Received: by ftp.ardi.com (Smail3.1.29.1 #3) id m0tYA6J-0007q3a; Fri, 5 Jan 96 04:14 MST Sender: owner-paper Received: from ardi.com by ftp.ardi.com (Smail3.1.29.1 #3) id m0tYA5J-0007qBn; Fri, 5 Jan 96 04:13 MST Path: sloth.swcp.com!usenet From: Clifford T. Matthews <ctm@ardi.com> Newsgroups: comp.emulators.mac.executor Subject: Re: Error 47 in PBHDelete (?) Date: 05 Jan 1996 03:43:36 -0700 Organization: ARDI Lines: 46 Message-ID: <ufd98ysyzt.fsf@ftp.ardi.com> References: <117cc$53823.2ce@newshub> NNTP-Posting-Host: ftp.ardi.com In-reply-to: TPub's message of Mon, 01 Jan 1996 13:56:35 GMT X-Newsreader: Gnus v5.0 To: executor@ardi.com X-MailNews-Gateway: From newsgroup comp.emulators.mac.executor Sender: owner-paper@ardi.com Precedence: bulk >>>>> "C.S." == TPub <tameri@lightspeed.net> writes: In article <117cc$53823.2ce@newshub> TPub <tameri@lightspeed.net> writes: C.S.> I was deleting a folder a few minutes ago and was greeted C.S.> with a nasty error 47 message. Now, I cannot delete the C.S.> folder, and when I open it, it is a mirror of the "root" C.S.> folder of the volume. This is a very serious problem. Right now I'm "Mr. Filesystem", but I (along with Mat Hostetter and Bill Goldman) am about to leave for MACWORLD Expo. I've logged this bug and will be sending you more e-mail when I get back. C.S.> The error seems related to the ISAM used to mimic the Mac C.S.> file system. Do we need to plan a "Disk Doctor" for C.S.> Executor? (* Just Kiddin' *) Norton Disk Doctor 1.0 (for the Macintosh, of course) will work under Executor (if you enable direct disk access). We haven't been able to get NDD 2.0 to work, and we're aware that there is an even newer version out there. Getting them to work under Executor is fairly high priority although it probably won't happen until after Executor 2 is released. C.S.> As a side note: is a utility planned by anyone out there to C.S.> "expand" and "contract" Executor volumes? Creating a second C.S.> volume file, moving data, then deleting the first isn't C.S.> always an option of some media -- there just isn't enough C.S.> space. We certainly would like to do this. Exactly *when* we'll be able to do it we don't know. C.S.> BTW: I never did get Crystal Quest to work in p9, but it C.S.> does run properly under p8. It locks solid under p9. Hmmm... I hope it runs better in 1.99q. If not, we'll need to get more information from you about this, too. Of course CQ isn't nearly as important as a filesystem bug that can lead to corruption. Always back up your data! C.S.> - C. S. Wyatt --Cliff ctm@ardi.com From owner-paper Fri Jan 5 04:39:56 1996 Return-Path: <owner-paper> Received: by ftp.ardi.com (Smail3.1.29.1 #3) id m0tYAUx-0007qBa; Fri, 5 Jan 96 04:39 MST Sender: owner-paper Received: from tide10.microsoft.com by ftp.ardi.com with smtp (Smail3.1.29.1 #3) id m0tYAUN-0007q3C; Fri, 5 Jan 96 04:39 MST Received: by tide10.microsoft.com; id DAA09342; Fri, 5 Jan 1996 03:40:12 -0800 Received: from unknown(157.54.17.74) by tide10.microsoft.com via smap (g3.0.3) id xma009326; Fri, 5 Jan 96 03:39:52 -0800 Received: from xnet2 (xnet2.microsoft.com [157.54.17.205]) by imail2.microsoft.com (8.7.1/8.7.1) with SMTP id DAA24194; Fri, 5 Jan 1996 03:29:44 -0800 (PST) X-Received: from NAX-01-HUB by xnet2 with recvsmtp; Fri, 5 Jan 1996 03:26:47 -0800 X-Received: from eux-02-hub by nax-01-hub with receive; Fri, 5 Jan 1996 03:25:27 -0800 X-Received: from lul-02-msg by eux-02-hub with receive; Fri, 5 Jan 1996 11:18:40 +0000 X-MSMail-Identification-du-message: 7BE00E8C X-MSMail-ID-de-la-conversation: 7BE00E8C From: Florent Pitoun (Itl Temp) <i-flopit@microsoft.com> To: executor@ardi.com, owner-paper@ardi.com Date: Fri, 5 Jan 96 12:17:23 PST Subject: RE: Resedit works on Executor 1.99p9, (finally)... X-MsXMTID: lul-02-msg960105111819MTP[01.52.00]000000f7-875 Message-Id: lul-02-msg960105111819MTP[01.52.00]000000f7-875 Sender: owner-paper@ardi.com Precedence: bulk Which version of resedit? Florent ---------- De: Merlyn Oppenheim <merlyn@dkunix.dkeep.com> A: <executor@ardi.com> Objet: Resedit works on Executor 1.99p9, (finally)... Date: jeudi 4 janvier 1996 17:16 just though you might like to know...Merlyn@dkmail.dkeep.com 
