On Tue, 12 Jul 2016, Cameron Kaiser wrote: That was the ROM code, right? I'm curious about that, myself. I guess that it can all be software emulated. I suppose they could have created some kind of software mechanism to capture those calls and redirect them to a library which re-implemented them. I'm guessing the same would be needed for Quickdraw calls, but perhaps those weren't around in the 6.x days? I dunno. I've never seen one in the wild. The picture looks like an early NEC workstation to me. It's sad that they never made it work to any major degree. The more the merrier, I say. -Swift

On Wed, Jul 13, 2016 at 12:39 AM, Chris Hanson <cmhanson at eschatologist.net> wrote: Was there a pre-68K period in Mac development? -- Eric Christopherson

Yes, 6809: http://www.folklore.org -> search for 6809. Regards G?tz

On Jul 12, 2016, at 9:25 AM, Cameron Kaiser <spectre at floodgap.com> wrote: There?s nothing particularly special about the Mac Toolbox and Operating System per se. Pretty much anyone could have attempted to develop a clean-room workalike using the ?Inside Macintosh? books: - Implement each API described in the book to have the behavior described in the books, - Dispatching each one using the A-trap and taking parameters in the registers described in the books, - And affecting the globals as described in the books. Through System 6 there were only a few hundred and they were very well-understood, and even things like trap patching were (relatively) well managed by developers of large scale commercial apps because they had to work on everything from a Mac 512Ke to a Mac IIfx with tons of RAM and disk and multiple displays under MultiFinder or A/UX. Many companies actually developed portable workalikes to the System 6 APIs in order to port their applications from one platform to another. QuarkXPress even used to provide a Toolbox-workalike API as part of their Windows plug-in SDK. The real killer was System 7, which doubled or tripled the number of available APIs via the system, and did so with thorough integration and compatibility. Then there was the PowerPC transition and System 7.1.2. And then the API set grew enormously again with System 7.5? There was no way someone like NuTek could have kept up. There were companies that developed later Mac API compatibility suites. One of them was Altura, the same people who provided the Quick Help application that all of the MacOS programming docs switched to (in lieu of Symantec?s THINK Reference); they provided an API suite that you could use to port to Windows or UNIX and they had at least a minimally working version for OPENSTEP after Apple bought NeXT and before Apple announced Carbon. And of course Apple itself had a Mac API compatibility suite that was part of QuickTime for Windows. I know of companies that used it to actually port Mac applications to Windows, because it was fairly complete and licensing QuickTime for Windows for a commercial product wasn?t a hassle. -- Chris

On Tue, 12 Jul 2016, Liam Proven wrote: That's about what I'd expect. I wonder if it could crash as much as OS 8.1 on my Quadra 700. That's a tough act to follow. :-) IIRC, there was an alpha-quality liveCD for a while. I never could get that excited about NeXT, Objective C, or any of that Steve-Jobs-in-limbo kruft (and by extension GNUStep, either). I saw a Color Turbo slab for sale recently: http://denver.craigslist.org/sys/5677975263.html I passed. That machine is sweet, for what it is. However, like most hobbyists I tend to gravitate toward machines I actually used "back in the day". In the 90's I was a student, mostly. There was no-freakin-way I was going to afford a NeXT machine. They were prohibitively expensive (or at least that's my recollection): even more so than high-end Macs. Plus, back in the 1990's I met a couple of people who did own them, and they were *super-snobby* about it, which also turned me off. It's a bit like BMW owners today. I don't care if they put 1000 HP in them, even most of their sportscars ('cept the whacky hybrid) still looks to me like mom's car leaving the tennis courts at the country club to head out to a PTA meeting. I'm guessing I will never be a BMW fan or a NeXT bigot. GNUStep wants to clone their whole API and the UI, as you know. I wish them luck but it's nothing that exciting to me personally. It's interesting that you bring it up now that Linux is committing anti-UNIX heresy on a regular basis. Maybe GNUStep's future is now brighter? It's still very fiddly and immature the last time I looked at it, but in terms of the overall approach, it does appear to have some nice plumbing and backing-ideas. I'd rather see GNUStep succeed than GNOME or KDE (fantasy on my part), honestly. Those two are just hopeless chaos-impregnated hairballs with ridiculous dependency chains which are starting to pollute working/good/not-at-all-broken areas of the *OS* at this point. I've never liked either project (though I could almost stand GNOME for short periods in the early days). Then again, I'm not one of those "Linux world domination" types who want to somehow capture every user, no matter how low we have to set the bar to snag them. Google Android has shown that folks can (successfully) bastardize Linux/UNIX into something very weird, proprietary, custom, and no longer even resembling UNIX, much. So, now that this sort of blaspheming is normal, why not try to make a *decent* desktop OS from it, eh? Lord knows, Ubuntu is trying. Who knows, maybe Android will become that. I'll catch the screenshots... I'd rather not use an OS where soooooo many of the apps are pre-infected with some type of malware or does things behind the scenes I wouldn't approve of (yet the "store" claims they are "virus" free, eh?). Funny how they can redefine "virus" or "malware" as it suits them (ie.. corporate sponsors say it's safe? Oh, ohhhhhkay then, we don't mind if you steal an address book, log keystrokes, or secretly GPS track folks - just don't replicate). The countermeasures for these issues seem to me to be weak and ineffective, so far. I'm not sure I'll ever be able to trust commercial OS's or software at this point, no matter how much bling they cop. I still carry my Philips Xenium phone running Symbian (and lasting about 20 days before needing a charge). Tastes great, and less filling. I'd love to see a commercial phone OS project start with the mentality of the OpenBSD project. I'd be willing to try something like that! Features like totally secure defaults, zero trust for basically anyone or anything, secure OS protections that are difficult to override by silly apps, etc.. would be welcome. I have absolutely zero doubt that you are quite correct. If it took .000001% bit of market share away from them, they'd have a nuclear freak-out and figure out a way to hybridize ninjas with their corporate lawyers and send them out riding elephant sharks for vengeance. What would be hilarious (but again fantasy) is if ReactOS had a breakthrough in terms of functionality that got them very close (say 99% or better compat). Then if they sat on it for a while, getting it right before subsequently release it the genie would be out of the bottle. If it worked compatibility-wise even as well as XP or Win7, it'd be a hit and crimp the snot out of M$. Of course, they'd probably find some way to DCMA it out of existence. It just depends on how widespread the release got and how illegal it was to own it. It's impressive when it works. There are still a large number of applications that don't work, too. Even bread-and-butter apps like the latest Firefox often crash and burn. I have seen a few that are rock solid, Office 97, Winamp, and a few other "gold" (winehq) or better certified applications. I run RegexBuddy sometimes in Wine and as you say, no problems. That's not to say the WINE team isn't amazing. They are. It's just a tough slog. -Swift

On 12 July 2016 at 20:06, Swift Griggs <swiftgriggs at gmail.com> wrote: That was one of the things people didn't talk about in the classic days. I supported classic MacOS Macs up until the early noughties. They were horribly unstable. I embraced OS X early on, but some people hung on as long as possible, and others disliked OS X so much they switched to Windows. All the now-nostalgicized-over '80s OSes were pretty horribly unstable: Windows 3.x, and indeed 9x; Amiga OS; ST GEM; Acorn RISC OS. None had proper memory protection, few had preemptive multitasking or didn't do it well. It's why I switched to Windows NT 3.1 back in '93 and liked it. I did actually used to like Windows, an NT was a proper, solid, grown -up OS. In my experience, more stable than OS/2 >=2. Exactly. Wow! That's great value! If only I were on the same continent! True. 1980s for me. The expensive kit I couldn't afford were things like the Apple ][ and BBC Micro, or even a fully-tricked-out C64. Later the Amiga, ST and Mac. By the time I had some money, 2nd hand Acorn Archimedes were available for <GBP1K and that was, to quote my late friend Guty Kewney, "my first experience of Raw Computer Power". Yes they were. Part of a non-compete agreement between Steve Jobs & Apple Computer, you see. Well, they had reason. In their time they were /incredibly/ radical computers. Wouldn't know. I don't do cars. I like BMW bikes, though. Had an R80/7 with a sidecar for many years. Interesting. I think it's a hugely big deal, but it's too late, sadly. It's a million-to-one shot, I reckon. And similar comments to those re ReactOS and MS lawyers apply. Me too! GNOME 2 was all right. Best desktop of its time. GNOME 3's main role seems to be inspiring replacements for itself and providing some foundations for them. ;-) I think it's arguably happened, actually, in the form of your next paragraph. Well quite. Nah, not with Chrome OS etc. around. A fair point. I loved my Blackberry Passport -- its OS really is a better Android than Android. However, it just doesn't have the apps. BB10 is close, but proprietary. Perhaps Jolla comes closer? :-D Yep! I reckon a 64-bit version of Windows 2000, with support for the newer APIs -- to hell with Metro etc., just classic Win32, no DOS, no 16-bit, no 32-bit, pure legacy-free 64-bit with none of the accessories, UI changes etc. No "Active Desktop", no IE, just the Explorer and enough compatibility to add 3rd party replacements for everything from Notepad to Paint to Media Player. No sidebar, desktop gadgets, Start screen, Modern apps, nothing. There's a lot of free-to-download Windows components, from Media Player to IE. It just needs to be good enough to run them, it doesn't need to replicate them. Even so, I don't think it will ever happen. Exactly. Anything that runs WINE has native Firefox, surely, plus other browsers? Absolutely. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

On 14 July 2016 at 19:34, Fred Cisin <cisin at xenosoft.com> wrote: Ah yes. Now I live relatively close to Ukraine, I thought of getting one. But the company has shut down due to the war with Russia and they've gone up in price 10x over. :-( My mum had one. She demolished a gas station kiosk with it, then later, drove home from work, drove into the garage... right up to the back wall, trapping herself in the car as its door opened forwards and it had no reverse gear. :-D She sat there for a whole day until my dad got home from work and freed her. :-) I only had minutes on one, once, at a trade show decades back. :'( True, but does it matter? Oh, it's being tried: http://www.theverge.com/2016/1/6/10726986/remix-os-android-desktop-ces-2016 Long term, I think Google should find some way to converge ChromeOS and Android. Having 2 different Linux-based OSes seems redundant and a waste of effort. And there's an internal-only Linux server distro too, I hear. But they can afford the duplication of effort. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

Personally - I went through my larval phase under it - I'd cite VMS as a counterexample. Even today I think a lot of OSes would do well to learn from it. (Not that I think it's perfect, of course. But I do think it did some things better than most of what I see today.) /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

For its time, certainly. Even today, there are a few things a DECnet stack does better than an IP stack. I'm not sure I agree. The VMS command line I used sucked, but so did Unix shells of the time, and in many of the same ways. As for VMS HELP, I don't think the tool is all that much better; what is _much_ better is the documentation it contains. DEC documentation of the VMS era was _awesome_. Even today I rarely see it equaled, never mind bettered, in many ways. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On 14 July 2016 at 23:51, Peter Coghlan <cctalk at beyondthepale.ie> wrote: This! I learned VMS at uni in the mid-1980s. It was my first proper CLI -- before that, my computing experience consisted of ZX Spectrum, CBM PET and very briefly TI 99/4A. All of those had BASIC in ROM, so they weren't true command shells. The BBC Micro had a separate OS from its BASIC and did have a sort of CLI, later more completely separated off in RISC OS -- but I couldn't afford a BBC Micro and neither could my school. I still prefer the DOS/NT shell to Unix ones, to the horror, dismay and disgust of all my Unix-using friends. The Unix shell does all kinds of fancy stuff I never need, but it makes things I use a lot, like wildcard renames, much harder than on CMD.EXE. So, yes, I liked DCL and thought it was a pretty good -- if wordy -- shell. I don't suppose I remember much now but I preferred it to Unix -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

I've seen a lot of complaints about this over they years but I've never really understood the problem. I think wordier commands in a command procedure (VMS speak for what others might call a shell script, batch file or an exec) are easier to understand. When they are being typed at a command prompt, they can be abbreviated somewhat to avoid redundant typing although they will never be as short as in certain other operating systems. I guess it must irritate a lot of people though because it keeps coming up. I particularly like that items like dates/times have a standard form and they they work exactly the same with every command (unless the programmer just doesn't get the "VMS way" and works really hard to prevent it). I think that dates/times were done pretty well on VMS with the exception of a couple of blunders - not going further back than 1858 for the base date and not having the system manage time in UTC while allowing individual users to deal with time in whatever timezone they want to be in. While I like the way DCL processes individual commands, I think it is a bit weak when it comes to scripting command procedures and I would prefer to have something that processes commands like DCL but has facilities more like IBM's REXX for building command procedures (if that doesn't cause too much annoyance in to those on both sides of the DEC/IBM fence...) I never used MCR at all. Regards, Peter Coghlan.

On 14 July 2016 at 22:43, Mouse <mouse at rodents-montreal.org> wrote: HELP WOMBAT -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

On 15 July 2016 at 17:57, Paul Koning <paulkoning at comcast.net> wrote: :-) I spent /hours/ reading that. At first I was looking around for the hidden camera because I was convinced someone was playing a very sophisticated practical joke on me at work... Sounds great. I never saw a PLATO terminal. :-( Wish I had now! -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

Anent PLATO discussions, that reminds me: I discovered that I have a 360K (DSDD) floppy with (apparently) PLATO client software on it probably from the mid 1980s. Is the image of this of any interest to anyone? --Chuck

On 7/15/2016 9:29 PM, Jason T wrote: I so wish I could find my copy of the PLATO client for the Commodore 64 (yep, it actually existed, I did some of my Physics 107 labs on PLATO to get to the terminal server(?) at the home of PLATO, UIUC. Sigh, I miss that. -- Jim Brain brain at jbrain.com www.jbrain.com

On 7/16/2016 12:54 AM, Jason T wrote: It was interesting. It managed to squish the 512x512 into a 320x200 somehow. I had one of those cards as well. jlb31348 at usa.cso.uiuc.edu. 1989, I believe. It might have been 4000, 1000 might have been the UIUC general help line. It's been almost 30 years, the memory, she did not keep all of that info handy, as I never thought it would be useful. PLATO was such an interesting thing at the time at UIUC. It was considered "old" tech, and "HyperCard" for the Macs were the new way to do lessons. THose wooden cases and orange screens were an interesting sight. I always knew there was a huge community on the PLATO machines (University HIgh School in Urbana had them, as did other schools), but I never got into it. IRC and USENET demanded my time (feel free to search for jlb31348, I am sure there are plenty of horrid postings on comp.sys.cbm and sci.virtual-reality from my days there). Jim -- Jim Brain brain at jbrain.com www.jbrain.com

On Fri, 15 Jul 2016, Liam Proven wrote: I wish they'd had a few at schools I attended. I think someone on the list mentioned that PLATO content could be viewed on Apple hardware, too. The wikipedia article on it is very detailed. I've always liked the idea of a "full educational kit" meaning that someone creates a nearly comprehensive set of documents written stepwise journeyman's level of skill with as many other self-help/self-learning tools thrown in as could be possibly useful. From the description, PLATO seems to have embraced that idea at various points depending on who was writing content. Cool things about PLATO: * It had graphics, but ran on terminals! * It could do animations in the content * It supported speech synthesis. Blind folks want to play too! * Cool people were involved (NSF, Navy, Air Force, many scientists & engineers, Control Data, etc..) * It had a flight simulator! * It punched above it's CPU power for a i8080 * It was said to be easy to code for (TUTOR was the lang, sayeth wikipedia) * They had MUDs and other cool multi-user games, as well as "social media" (ie.. chat and multi-user applications). * Even way back when, they had touch screens! I'm sad I didn't get to learn physics 101 from one! However, my instructor for that class happened to be awesome, so maybe I should have said Linear Equations or Calc II. I had foreign unintelligible mealy-mouthed cut-rate TAs teaching those classes. Puh. I'd have taken an PLATO terminal ANY DAY over those guys since their content would have presumably been in the Queen's clear readable English. Nowadays you have Khan Academy (go Khan!) and other places that have some pretty fabulous courses and content. Not to mention big unis doing open-courses. I think both MIT and Stanford have them. I've downloaded books and materials from the MIT Open Courseware. I also like to take or at least skim courses on things I'm not familiar with aimed at kids. They make a lot fewer assumptions. Motivation I've got. 40 extra hours a week for classes at a brick and mortar school, I sadly do not have (unless I want to lose some serious sleep). So, anything that bootstraps my knowledge in an area in a complete but as-I-get-time fashion, I'm 100% on board with. I also keep old CBT CDROMs and instructional DVDs for various things. They might be old, but they often have more content or did a better job with the illustrations or animations than you get on the web. Learning is great fun to me. School, uhh, not as much. However, I know some people find the collaboration, a live instructor, and friends they make in the social atmosphere to be invaluable for their learning and enthusiasm (which is a learning amplifier, IMHO). I also have to admit that I did learn quite a bit in "labs" for classes I had, especially Astronomy classes. The labs were what kindled a sense of wonder in me. So, learning comes in a constellation of formats. I personally just like the ones that are self-driven the best at this point. I wonder what takes the place of things like PLATO nowadays. Probably a hodge-podge of PeeCee Windows apps and Adobe Flash/AIR apps, I'd guess. I'm not involved in any kind of formal education at this point, so I wouldn't know. -Swift

On Jul 15, 2016, at 10:03 AM, Swift Griggs <swiftgriggs at gmail.com> wrote: Graphics terminals were a thing that existed. It wasn?t just PLATO that used them. -- Chris

On 07/15/2016 05:47 PM, Paul Koning wrote: Surely you remember CDC IGS from the 70s. I loved watching the displays being drawn on those big radar CRT displays--one color while drawing and persisting in another. They were "terminals" of a sort, no? --Chuck

On 07/16/2016 10:34 AM, Paul Koning wrote: IGS (Interactive Graphics System) was the generic term for the Digigraphic 200-series stuff. The consoles came in various flavors, but the best ones were the big, flat-surface jobs. The display processor with these things was pretty large; someone once told me that it resembled a 1700 more than anything. A lot of fun to fool with. --Chuck

On Jul 15, 2016, at 9:34 PM, Eric Smith <spacewar at gmail.com> wrote: The PLATO IV terminals had a 512x512 addressable pixels, local charset memory (Font) and the ease and power of TUTOR to support them. It still amazes me how much work and fun we extracted from the limited cpu, memory, storage and communication bandwidth we had. Oh and those keyboards. Best damn ones I?ve ever used. Jerry

I was running a 3 node VAXcluster in the late 1980s. We had two 8550s and an 8820 connected via a CI star coupler to two HSC70 storage controllers and 24 RA81 drives; two upright tape (TU78s?) drives too. The drives were connected to both HSC70s in RAID 1 pairs. We had 11 pairs, a spare and a quorum disk for the VAXcluster. The environment was rock solid and ran for many years. We could do rolling VMS and application upgrades on the three nodes. A great production system. We even had an X25 based DECnet connection between Australia where the system was installed and the UK where our software company was based. On 14 July 2016 at 21:50, Swift Griggs <swiftgriggs at gmail.com> wrote: -- 4.4 > 5.4

If you think of "networking" as being "IP-based networking", yeah, probably. But there's a lot more to networking than just IP. Specifically, I was talking about DECnet, which was well done and integrated from the ground up, not glued on after the fact. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Let?s say UCX had some deficiencies (being polite) and was replaced with TCPIP Services for OpenVMS. This TCP/IP stack was based on the code from Tru64 Unix (aka Digital Unix aka OSF/1) and used what was known as the basket to map the OpenVMS API to Tru64 and vice versa. Disclaimer: I used to work for HP and was an OpenVMS Ambassador so might be slightly biased Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au Melbourne | "If soccer was meant to be played in the Australia | air, the sky would be painted green"

I think it is only semi-proprietary. I've seen open documentation that at the time (I don't think I have it handy now) I thought was sufficient to write an independent implementation, both for Ethernet and for serial lines. However, IIRC it also has a fairly small hard limit on the number of hosts it supports. I don't remember exactly what the limit is; different memories are handing me 10, 12, and 16 bits as the address size, but even the highest of those is sufficient for at most a large corporation. (Maybe it was 6 bits of area number and 10 bits of host number within each area? I'm sure someone here knows.) Perhaps if DEC had enlarged the address space (somewhat a la the IPv4->IPv6 change) and released open-source implementations, it might have been a contender. For all I know maybe they've even done that, but now it's much too late to seriously challenge IP's hegemony. But the real shining star of DECnet/VMS was not the protocols, but the ground-up integration into the OS. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

Correct. 16 bits total in Phase IV (up from 8 bits in Phase II and III). Then again, with NAT ("hidden areas") that worked acceptably well even for the largest DECNet (the one at Digital). Keep in mind that DECnet was designed as a network for an organization, not as an internet. DECnet did increase the address space, with Phase V where the address is variable length up to 20 bytes. The difficulty is that it was all based on OSI, with all the international standards bureaucracy that implied. And by that time, TCP/IP had become a viable competitor, which was "good enough" (32 bit addresses) and sufficiently much simpler and more nimble that it came out the winner. Well said. paul

True, OSI application layer protocols tended to be elephants. But one should not confuse those with the lower layers. You can run X.400 on top of TCP or DECnet NSP (DEC did exactly the latter, in a commercially successful product despised by the engineers using the internal network), and similarly you can perfectly well run non-OSI applications on top of TP4. For example, it would be trivial to run http, or iscsi, over DECnet. It hasn't been done as far as I know, but technically it would be a no-brainer. In fact, a lot of application protocols (iscsi for example) are easier on DECnet because NSP (and TP4) have packet boundaries while TCP does not (though SCTP does). paul

On 15/07/16 14:49, Swift Griggs wrote: VMS came with DECnet built-in (although you had to license it). If you wanted TCP/IP there was UCX, which you had to install separately. The other TCP/IP stacks came from 3rd party vendors. That's why there were multiple implementations of TCP/IP for VMS. The specs were (and are) freely available. (I'm not 100% sure that they were free-as-in-beer back then, but they are now). There was at least one implementation for Linux and (I think ...) another for Solaris. cisco also supported DECnet in some of their switches. Seldom used *now*. All the VMS systems I used commercially back then made use of DECnet. IPv4 and IPv6 are also only loosely related. At least the DECnet phases were sequentially numbered :-) (I'm assuming that Phase II existed at some point before Phase III, which definitely did exist. I also assume that Phase I only acquired that designation once Phase II appeared!) Antonio -- Antonio Carlini arcarlini at iee.org

By the way: starting with Phase III, DEC adopted "one phase back" compatibility. A Phase III node could talk to Phase II; a Phase IV node could talk to Phase III, all documented clearly in the specifications. (Two phases back wasn't described or aimed for, though it is not that hard; my DECnet/Python does Phase IV but talks to Phase II.) On the other hand, Phase II is not compatible with Phase I; the packet formats are significantly and it's clear that no attempt was made to deliver compatibility. I don't know why not, or why that changed later. (Before my time...) paul

I'd always heard that. But recently I found Phase I documents, which include protocol specifications of a sort, sufficient to tell that it wouldn't be compatible with Phase II and couldn't readily be made to be. (In particular, NSP works rather differently.) And that document was for a PDP-8 OS. Possibly it was built but not shipped, or designed but not built. The document has the look of r a finished product manual, though. paul

So does that mean that RTS/8 DECnet Phase I was built but not shipped? Or shipped but not supported? The document I referred to is a full manual "RTS/8 DECNET/8 User's Guide, Order No. AA-5184A-TA". A note at the start says "converted from scanned text 1-Jun-1996" and just below that "First printing, February 1977". Chapter 6 is a fairly detained description of protocol message formats, which look vaguely like NSP as we know it but only vaguely. paul

Swift said: As a comp sci student I loved using VMS on our 11/780s at Uni, from first year through final year where we also had the use of a Gould PN6080 UNIX mini. (Aside - the Gould had one good drive, one flaky. The OS and staff accounts were on one, student accounts and /tmp on the other. Guess which :) On the teaching VAX, I vaguely recall one time just after the computing department had a new version of the OS installed, I logged in and I typed '&' (or something) on a line by itself and the DCL shell crashed and went back to login. That got patched pretty quick. Another humorous thing was certain faculties such as Statistics or Economics would hand out (apart from an account for each student) a common account that was locked into a DCL menu of for instance stats applications, that had a minimal quota and priveleges and anyone in the course could use to check terminal availability and print or submit job completions and that sort of thing. With these accounts it was possible to break out of the menu to the DCL shell, and as it was an anonymous account do (from hazy memory) something along the lines of EDIT/NOJOURNAL [SYS$SYSTEM]password.dat or something similar, and presto although you couldn't edit it or even see it, it would be held open and any attempt for anyone to log in anywhere would get some message that the password file was locked by another user. I er saw it done by a friend :) Apart from that, students would write crazy long DCL scripts that would find out whether their friends were logged in somewhere on campus, and that sort of thing. No matter that it took ages to execute and used up our meagre student account CPU-seconds quota and log us out! So we just logged in again and got another few CPU seconds. The messaging command (can't recall what it was - phone?) was great and lots of fun to use. Of course geek guys would use it to send messages to girls they could see at other terminals, offering to help! I recall using EDIT/EDT and really loved it, none of our student terminals (Telerays?, Hazeltines, LSI, Wyse, any other cheap beaten-up terminals the Uni owned) ever had the mysterious GOLD key though, and it wasn't till decades later I saw a real DEC keyboard with that key. I felt disappointed because it was actually just yellow and not really gold at all, not even painted. Other times I used to edit my comp sci and stats assignments in line mode on the DECwriter IIIs and Teletype 43s which most students avoided like the plague, preferring to use EDT in full-screen mode on a glass terminal. Being comfortable with line mode editing was very convenient for me if I happened to arrive late to a terminal room when assignments were nearly due. And now I have one of those cute little baby VAXen, the smallest VAX ever made, a 4000 VLC from an eBay impulse purchase. I have not powered it up yet but someday I will and am hoping it works and has VMS on it. It might even jog a few more fond memories (^_^) Steve.

On Thu, Jul 14, 2016 at 4:50 PM, Swift Griggs <swiftgriggs at gmail.com> wrote: It was a huge deal in the late 80s and into the 90s. I was on both sides, so mostly, I watched. I got my start with VMS a few months before I touched UNIX - same hardware - VAX-11/750. I've written device drivers, system utilities, and application code for both. VMS was very good to me from 1984-1994, and I did a bit more with it from 1997-2003, then nothing commercially since then. UNIX (and by extension Linux) has been good to me the entire time since 1985. If I have choice, I'll grab something UNIXy to do my work on - I'm not particular as to flavor. I think TCP networking on VMS is a bit of a bodge, but back when I used it every day in the 1980s, we didn't _have_ any Ethernet interfaces in the entire company - *everything* we did was via sync and async serial. How well do you think it would go if all you had was SLIP and PPP? We did a lot. Yes, other people had high-speed networking and VAX clusters, etc. We did not. Not even our VAXen running UNIX. All serial, all day. We still got a lot done. I found certain aspects of DCL to be quirky, even if I did learn it before I touched a UNIX shell. That said, it was easier (to me) to write full-on apps and utilities in DCL than sh or csh. It would be a fairer comparison to develop a complex app in Perl vs DCL (Perl would win, but it has a lot going for it). I even completely automated our build process (formerly a full-time engineer's job, but as the company shrank, we couldn't afford to have someone who was, essentially just a build master)... source code control, pulling code based on which product it was for, compiling it (without "make"), linking it locally, sending the objects over to a machine running another version of VMS, linking it there, pulling all the text objects and executables into two tape-build repositories and cutting magtape for distribution - all in DCL. I literally turned a fulltime job into a script. All you had to type was "$ RELEASE <protocol> <hardware bus>" and it would pull everything, auto-increment the version number, inject it into the code, build everything and tell you it was time to make tapes (8 hours later!) I'm sure it's possible to do all of that in csh, but even now, I wouldn't want to be the one to build that. The regularity and predictability of args and options is definitely a strength in DCL. Args are entire words, not letters which change from app to app. Here's a trivia question: which letters are _not_ valid arguments to 'ls'? I know one off the top of my head but not the others. Next thing - how about those args to 'dd'? Crazy. Now how about 'tar'... etc., etc. I use this stuff every day, but I have internalized a massive amount of UNIX trivia to be able to do so. VMS requires far less random factoid knowledge to get stuff done on the command line. There's a system command line parser, and it helps with the consistency. VMS HELP is also awesome. I use man pages - they are good if you already know how things work and are just trying to remember is it '-p' here, or '-a' or what? Much stronger. There are dozens of privileges you can grant so someone can do their job and not overstep things. UNIX says, "all or nothing. Don't screw up." [ other strengths deleted for brevity ] There was DECUS back in the day - a very strong source of sharable software going back well over a decade before there was any VMS. Consequently, there was less freeware available via Usenet. OTOH, I learned a *lot* porting utilities and games from comp.sources.unix and comp.sources.games to VMS. Some things were a lot harder than others. Meh. There are looping directives, but they are not as C-like as sh/bash/csh/ksh... Very true, but back in the day, you had to pay ~$30,000 to get the UNIX source. I have the papers to prove it (Software Results had a System III and System V source license, which I ended up with when the company folded). You could also pay DEC for VMS source. No idea how much that was (not cheap). For some perspective, you were running these OSes on machines that cost $300,000 in 1977 (11/780), $120,000 in 1980 (11/750), $40,000 in 1982 (11/730), $10,000 in 1985 (MicroVAX)... (newer, smaller and slower, at least the models we could afford). No offense taken, but I'm just sharing my own experiences as someone who used both, heavily, every day for most of 1984-1994 on 80s hardware, and still use UNIX and UNIX-y things all day every day. In the end, it did thrive where VMS was something of a dead-end. Fun ride while it lasted, though. -ethan (Some prices from memory, some checked against resources like http://www.jcmit.com/cpu-performance.htm ) ~

On Fri, Jul 15, 2016 at 10:44 AM, Swift Griggs <swiftgriggs at gmail.com> wrote: Indeed. As you've seen, I use both. No need to be all "Commodore vs Atari" about it. ;-) I mean vs ethernet-type networking. The physical layer stuff has fewer variants to worry about with Ethernet vs serial (3mbps vs 10mbps vs 100mbps, and 10Base5 vs 10Base2 vs 10BaseT vs flavors of fiber as opposed to all the parameters one has to match up to get any two machines talking over a serial link). Where this matters is that all our modern gear was developed in an environment where nearly everything being transported across it is TCP/IP. Try pushing DDCMP over the wire. ISTR there's now some TCP wrappers to get gear to be willing to handle these packets, but that just adds to the complexity and frustration. With serial DDCMP, we just hooked up two sync serial ports up with a modem eliminator (which provides the baud-rate clocking for both hosts) and it "just works" (since there are few options to configure at that point). All the configuration is a layer or two up as you set up the logical nodes in your network. Entirely unlike TCP/IP and Unix networking in terms of workflow and type/quantity. This is not a "A is better than B" argument - it's just some descriptions of the elements of the process and how they are different. Sure. Absolutely no argument. Just pointing out that comparing DCL to shell isn't exactly apples-to-apples either. If anything, measured in arbitrary units, DCL is a half-step over shell scripting and a half-step below Perl (etc.) scripting. Have you ever seen a string of ''' used to dereference DCL args? Definitely the hardest thing about getting a working complex DCL script. I don't mean file permissions, I mean system privileges. Some UNIX filesystems have ACLs (VMS has _very_ well developed ACLs, but that's not what I mean). I mean "I am root" or "I am not root" in UNIX land becomes, "what system object/resource do you wish to access? Read or write? Do you have one of the following privileges: NETMBX, TMPMBX, GROUP, GRPPRV, ACNT, ALLSPOOL, BUGCHK, EXQUOTA, GRPNAM, PRMCEB, PRMGBL, PRMMBX, SHMEM,ALTPRI, AUDIT, OPER, PSWAPM, SECURITY, SYSLCK, WORLD,DIAGNOSE, IMPORT, MOUNT, SYSGBL, VOLPRO, READALL,BYPASS, CMEXEC, CMKRNL, DETACH, DOWNGRADE, LOG_IO, PFNMAP, PHY_IO, READALL, SETPRV, SHARE, SYSNAM, SYSPRV, UPGRADE? Want to mount a disk? In Unix, a user is told "must be root". In VMS, you need MOUNT. You can give someone MOUNT and none of the other privs, meaning this user can mount disks or tapes but not necessarily read physical memory or bypass file permissions or write to device registers or any of the other privileged tasks. It's not all-or-nothing; you grant the level of access required and no more. (http://www.mi.infn.it/~calcolo/OpenVMS/ssb71/6015/6017p014.htm#vms_privileg…) Yes. That was one I just dodged. If stuff I was porting did a fork(), I just found something else to port instead. The workarounds, as you point out, are non-trivial and don't map 1:1 to what fork() does. -ethan

*tcsh*, yes. I now find it very difficult to use vanilla csh, even though (being a product of the University of California) that was the first shell I ever used as an undergraduate. TMTOWTDI. (Actually having written full apps in Perl.) ObOnTopic: I've always found DCL too damn wordy, but I appreciate its precision. I keep a VAXstation 3100 around just to remind myself "how the other half live." -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- He who Laughs, Lasts. ------------------------------------------------------

On 14 July 2016 at 22:50, Swift Griggs <swiftgriggs at gmail.com> wrote: I am no VMS expert. I used it, I liked it, I did very basic sysadmin on VAXen, but I've never brought up a machine from bare metal, for instance. (OK, once, kinda, on SIMH.) But that sounds like a very fair summary, perhaps the best I've seen. I'm hoping that VSI actually manage to rectify some of these. A modern x86-64 port, for generic hardware, with the GUI and everything all thrown in, *no* extra premium-charged anything, and perhaps an enhanced POSIX environment with some FOSS tools to facilitate porting stuff over from Linux. And it needs to be priced very very competitively, to make it cheaper than Windows Server on VMware at the very least. I'm not confident of its chances, though. Apple's OS X Server was a very solid product, keenly priced (0 cost user licences), and with excellent functionality and admin compared to Linux -- but nobody much used it and now it's almost forgotten, a sideline. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

On 07/14/16 12:42 PM, Swift Griggs wrote: I'm missing something here. Although most did/are using the Apple supplied GUI/Aqua, it wasn't a requirement. I have/run OpenWindows (compiled for OS X/PPC), and also, although mostly for fun, have a copy of the Mosaic web browser, also compiled for OS X/PPC. Aside from the Netinfo directory server, from a basic level, you can pretty much do & run anything you would on Solaris, Unix, *BSD or Lunix. What OS X didn't ship with wasn't too hard to compile on my own. In defense of OS/2, I went from straight DOS to OS/2 1.3. I was taking a lot of college programming classes, and in Assembly language specifically, I found any number of ways to blow things up and loose my work. OS/2 truly provided a "better DOS than DOS", and I could blow up a DOS session with my Assembly code and go right on working. Applications are/were a long story on OS/2, that I could write volumes on, but in short, if you wanted to play games, DOS and later, Windows was the place to be. Or the more 2000+ updated answer, on a game console. OTOH, how many word processors/spreadsheets/presentation programs does one need per OS? From a technical perspective, the only big problem I had with OS/2, back in the 1990's, was the single thread input queue on the new OOUI, WPS (Work Place Shell). OS/2 is now sold under the name "eComStation" and boots from JFS2 volumes. In summary, back in the early 1990's, I moved to OS/2. I didn't do it to get some application I needed, I moved for stability in the Wintel world. And for me, it did a great job. Jerry

Thanks for the comments, it's always educational to get the viewpoints and experiences from others, on items that are "shared ground". I didn't mean to come off like an OS/2 fanatic. I started using OS/2 around 1990, early 1991 at the latest, and short of Unix (I wasn't a Unix fanatic at the time, although I was coming up to speed), I still judge OS/2 to be one of the better x86 options for the early and mid 1990's. Its a given here that you looked at the software you wanted to run, then purchased the appropriate hardware accordingly. Thanks for the reminder on the Arca Noae, I'm sure I had read that previously, then just selectively chose to drop it from memory. I haven't used OS/2, or its derivatives, exclusively on a day-to-day basis since probably 1997 or 1998 at the latest. On 07/14/16 04:53 PM, Swift Griggs wrote:

I guess I am glad that someone getting something positive from windows. I have never viewed it as any more than a virus distribution system with a poorly written GUI front end. Jerry On 07/15/16 12:15 PM, Liam Proven wrote:

windows 95 - yea, even bill gates stated that windows 95 was the pinnacle. ease of installation - maybe due to the fact that the bulk, if not all of us here are experienced users, I've never understood the belly-aching concerning installation. Not for DOS/windows, not for OS/2, not for BSD, not for Linux, not for Solaris. Specifically when you are giving the installer the entire disk for the OS as a new system install. Just grab the disk then go. Other settings, like network, even if it is dhcp, have to be added somewhere, be it during the install or after the fact. OS/2 vs the windows GUI - sorry, but the best that anyone is going to be able to convince me on here is personal preference. Its a GUI on top of the OS where end users double click icons. Aside from the single thread input queue on early WPS, the sole advantage I ever saw that windows had over OS/2 was that early on, the *.ini files were text based on windows vs binary on OS/2. At some point, ms followed IBM and moved to binary *.ini files. I don't remember at what version. There were nice GUI based applications (3rd party) for editing OS/2 *.ini files, but it was never as nice as having actual ASCII text based files. Jerry On 07/17/16 09:48 AM, Liam Proven wrote:

On 14 July 2016 at 22:51, Jerry Kemp <other at oryx.us> wrote: *Blink* Really? I did not think it was possible to boot OS X in multiuser mode _without_ loading Aqua and the desktop. Am I wrong? Darwin, maybe, but AFAIK Darwin isn't maintained any more, is it? Interesting. I didn't do much programming on OS/2, more on plain old DOS, but I could readily crash my OS/2 2 home PC with Fractint. Its fancy video modes could instantly cause OS/2 to throw an exception and halt. Hmmm. I take your point. I was never a gamer and Win3 apps ran great on OS/2 2, IME. :-) Variety is the spice of life? Indeed. And honestly WPS was really not all that as a shell. I place it down there with Amiga Intuition in its clunkiness. Classic MacOS, OS X and Win9x were all slicker and more capable IMHO. Indeed. I've tried it. It's just as much of a PITA to install as it was 20y ago. :-( I went from OS/2 2 to the beta of Win95, and then, later, to NT 4. At work, I used NT 3 -- for me, 3.51 was a classic version. No fancy UI but solid and capable. By modern standards, fast, too. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)

On Fri, 15 Jul 2016, Liam Proven wrote: Woot! The benefits of working with small clients over time. Well, if you ever put hands on another M68k, you might give it a shot. The key is to have an extra partition to setup with a BSD disklabel et al. If you have enough space (or a spare disk) it's pretty darn straightforward. It loads using a MacOS based loader program, so you don't have to ditch MacOS, either. However, the install is pretty raw (I like it, but I have a feeling you wouldn't). However, it's nowhere near as raw as, say, OpenBSD's installer. If you ever happen to install OpenBSD, Liam, please have a video camera rolling. I will be able to get all the choice British curse-phrases in one go that way. Also, just as an aside, your ex-roomy who told you that you weren't liking parts of UNIX because you weren't a dyed-in-the-wool coder (not to say you aren't smart or technical or can't do what you need to do with coding) was right. It's a programmers OS and it panders to coders and admins, others will be grousing about weird things they don't need and don't see a reason for, items being over-minimized, too spartan, or downright bizzare and not enough in the way of well-integrated features for users with other goals besides coding. Fully 100% agree with that dude, and I totally acknowledge that there is a rusty tetanus side of that double edged sword. That's why I still dabble with the darkside and play with GUI-focused OSes, too. It's a whole different feel. When I want to code, I plant myself in front of NetBSD or FreeBSD. When I want to record/compose a song, I break out an SGI, Amiga, or maybe someday a Mac (I got a fancy audio rig for my 68k Quadra recently). Hehe, I ran MorphOS, too. It was fun for a while, but I can't really handle a proprietary OS on a such a small scale. I got one way later, too. Well past when they were new/prime. I have the exact same feeling. For me SGIs were the biggest lust-target because I actually had played with them long enough to know what I was really missing (and I was younger and all that happy stuff). In my experience using tools like "ReEFIt" make multi-booting OSX and *ix or BSD on a Macs way easy, but yeah, they don't need much to "justwork" nowadays. My experience with it is less than 6 months old. Without Macosgarden I'd have never got the chance because finding legit disk for it is *hard* if you want 3.1. I had all manner of weird install problems because I was doing it on a SCSI2SD that isn't an Apple disk so of course Disk tools was pissed. The disk tools under A/UX would play nice, actually, but I ended up having to do all kinds of CLI jiggery pokery, manually creating file systems and what not from an emergency shell, to get A/UX to give up and install on the darn thing. It was damn weird (in a cool and unique way) once I got it working. and I dd'd off the install images and boot record off the MicroSD card once it had finished. I found that they more or less worked with Shoebill, at that point, too. I wouldn't call myself a master, but definitely an experienced power-user. I did quite a bit of coding using 386|VMM and other such things with mostly Borland tools. The thing I miss most about DOS was it's "standalone" mentality. You want to backup your word processor ? Zip the directory. You want to backup Deluxe Paint IIe? Zip the directory. You want to backup Lotus 1-2-3? Zip the directory. Everyone took a really long drag from the dynamic library joint and passed it around in the 90's, too. I took a hit, too, and I get that there are many advantages to them, but the big DISadvantage is now many binaries become version-specific to a library that may get deprecated in subsequent releases. On DOS, that wasn't a problem. Just keep running the old one. Sure you can still compile (most) things statically or include old libraries, but it's seldom done, fiddly for users, and oft overlooked. I often lament how most apps now want "merge" with your OS not simply run on their own in a super-self-contained way DOS apps did. I also tend to hate registries (not just on Windows, I despise AIX's ODM, too). DOS had no such BS, and it's one of the things I liked about it. ME was the one that pissed me off the most and seemed to understand my hatred and return it in kind. That OS seemed to me to crash for the smallest and most trivial of reasons. Windows didn't seem to get stable until the 3.x branch of NT matured into W2k in my opinion. However you are WAY more well versed in the M$ black arts, so I believe you when you say you could make it behave. Same here. One man's trash is another man's treasure. Whoa. Czech! I love that place. I was there in my 20's for a while and I have never seen so many beautiful women in my life (and they will actually talk to you occasionally, even if you aren't Fabio). The people were super friendly and welcoming to me, too. I'll never forget it. Of course it was pre-USA-terror-freakout days when folks in Europe loved Americans. :-) Nowadays when I go to Europe I pray people will think I'm from Canada. I've had some bad experiences when they found out otherwise :-( XFCE is a pretty solid workhorse that doesn't do to shabby at staying thin. I'm guessing you are going to be correct, and I try to keep that in mind as I bash on it. :-) To me it's a mole/privacy-invader, to others is a stepladder out of poverty. So, they win the relative importance comparison, for sure. I just wish we could both be satisfied. Well, riding on trains a lot and seeing people constantly on the phones, I see essentially three archetypes of glued-to-their-phones: 1. Young women (and men to a slightly lesser extent) who can't overcome their social drive and are just plain addicted to checking every possible social media channel (SMS, Facebook, Twitter, Snapchat, Instagram, rinse-repeat). That's pretty much all the do. I'm not sure that communication is beneficial or just plain compulsive/addictive behavior. 2. The I-have-a-million-apps-and-feel-empowered-by-technology hipster. This is the guy with not only his smartphone out. He's got the Apple Watch, two tablets, and an (unused) laptop in his bag behind him. He feels like a ninja because in his world, no matter what, "there is an app for that". He's wrong of course, and I'll eat him if we are ever stranded in the wilderness. 3. Business-weasels who are surreptitiously glancing at the other weasels to see if they have a more *expensive* phone, laptop, suit, or watch. If not, all is well. Now, he can settle in and try to figure out how to get this malware off his phone he got from surfing porn sites in Chrome from his phone. All humor and cynicism aside, I'm still on the fence when it comes to the benefit of phones in the first-world. The benefits come with some fairly heavy drawbacks, too. No seriously, I found it. It's in the Wikipedia page for ChromeOS. He doesn't sound super-committed, though. ha! That sounds about right. TBH, me too. However, the damn thing better run NetBSD. :-) Ohh, I just had an idea... I wonder if they can be hacked to run RISCOS. That would be really cool. I'll have to Google it. They are mostly ARM boxes, IIRC. It's true. The panoply of online services offering for free things that would have cost thousands (or been impossible) 15 years ago is staggering. Ha! Yep, I've read it. Good reference: right on point. I'd consider it if I hadn't just dropped $150 to buy a buddy's used Punkt phone. If you ain't sold it by this time next year, hit me up again ! :-) I don't use mine for weeks at a time. When it rings it's never a good sign (it's work, someone has died, or someone is wanting something from me I'm not going to be happy about). I probably would do the same, but I've always got that little bird on my shoulder saying "They are watching you and trying to screw you." I can't feel comfortable in what I consider to be such a user-hostile environment. So, I'm that Luddite on the train where there are 90 people in the train car and nobody under 60 is doing anything but staring at their phone... but me. I'm reading a some Stevens hardcover book or a trashy fantasy or SF novel. I might be tempted to use a laptop with a wireless 4G data card in there, but it's expensive and I get plenty of time on the metal anyway. Reading when I'm on the console helps me sharpen my mind (and it always needs sharpening, hehe). You'd have fun ridiculing my desk, then. I still have a Rolodex (paper one), a (very) old HP calculator, an At-a-glance paper calendar, and in my car I have a small bag with 4-5 books of maps. Yes, paper maps. In my pocket are a rinky-dink phone (as discussed), a 3" folding knife with screwdriver, a Fischer Space Pen, my leather wallet, and usually 4-5 3x5 notecards in my opposite pocket. Also, I often carry an Sony NEX-7 digital camera, if I have my pack. Rarely do I carry a "real" computer et al. It's probably because I have a different kind of job and travel needs versus you. You sound like you bounce around a lot. I have a pretty static routine these days. -Swift

On Tue, Jul 12, 2016 at 11:51 AM, Liam Proven <lproven at gmail.com> wrote: I assume you mean a distro that has GNUstep as its default UI the way Unity, GNOME, KDE, MINT, Cinammon, etc. do in various distros. Actually, in the early 2000s there were at least two embryonic distros I knew of: LinuxSTEP and Simply GNUstep. I followed LS a little closely for a while, but I don't remember much about the approach it took. I'm not sure it ever got so far as to really include GUI apps. I seem to remember that it took a less FHS-centric approach to directory layout, but it was probably less radical than things like the Bogolinux layout and more like the NeXT one. I definitely remember though that they didn't want to just through a GNUstep skin on the day's equivalent of Ubuntu (IIRC the distro was rolled independently of any other distro, except maybe Linux From Scratch, if that counts). Simply GNUstep I really don't know anything about. There just have never been very many GUI apps using GNUstep. In my experience, too, the actual experience of using them has been quite buggy, except for a few smaller apps (smaller = fewer features; I guess that means less to get wrong). Plus the interaction of GNUstep's GUI framework itself was, at least at the time, very dependent on the WindowMaker window manager, which was pretty much a separate product with its own set of problems (e.g. it's written in plain C using libraries that sort of mimicked the NeXT look and feel much the way GNUstep's own AppKit does, but with *no code shared* between the two). I just recently switched backed to WindowMaker and a mix & match of xterm, Firefox, Chrome, and the occasional Gtk+/GNOME or Qt/KDE app. I'd like something "pure" like an all-GNUstep system, but it's just not happening. -- Eric Christopherson

On Tue, 12 Jul 2016, sub at hoffart.de wrote: Quite right. I should have been more clear about both rather than use the term "emulator" which I already knew was pretty loaded with expectations. However, my point was only that if Executor could do it, so could someone like Nutek. However, from reading about the hardware, it sounds like it had an actual '030 in there. Well, WINE is probably a closer analog to Executor than ReactOS. However, again the point wasn't to define "emulation" but to say that re-implementing an OS isn't easy and has a spotty history of results. That's all. I agree with you that it's not impossible to do right. Just hard. I understand. I've used many such API-rigs myself under both NetBSD and Linux. I'm not sure what you're referring to here: binary emulation, POSIX compliance, Single UNIX Specification, iBCS emulation, etc... Personally, I've seen NetBSD "emulate" (for lack of a better term) a lot more commercial UNIX systems than Linux ever has. I've also had a lot more luck cross-compiling to other OSs which has some overlap with the whole API-emulation topic. However, in some of those cases you need to fetch libraries from the original dist to make it work right and that's probably "cheating" to most people who want to get froggy about the "emulation" term. It sounds interesting. I'm familiar with Firebee and MiNT, though this sounds a tad different. I'm not expert on STs, but I think they are neat. I'll have to check it out. -Swift