mayer@hplabsz.HPL.HP.COM (Niels Mayer) (10/21/88)
Could someone who has used the NeXT mailer comment on it's functionality. The pictures I've seen of their mail program makes it look like a pretty standard mailer. There seemed to be a bitmap with a face on it that perhaps shows the face of the person sending the mail (like Marshall Rose's extensions to mh for systems running face-server software + X windows). The buttonbox icons didn't indicate to me that the mailer has any particularly elaborate filing and searching capabilities. If it does, I'd love to hear about 'em. And what of voice mail. How does that work? Is voice data compressed, uuencoded, and attached to the mail file sent across the networks? I've heard rumours that there are some connections between the Andrew project at CMU and NeXT. In particular, that NextStep is layered ontop of something akin to the andrew toolkit. How true is this, and to what extent will NeXT software be able to exploit Andrew-like concepts such as graphical/voice/musical/active insets in documents. Is the NeXT mailer in any way akin to the Andrew Message System? Finally, does anybody have a line on CSCW (aka "groupware") software being prepared for the NeXT system? Colab-like software to allow groups to meet (real time) and work on design and idea-exploration through a bunch of networked NeXT computers. Extensions on the mail system to allow for structured messages to be created and automatically processed (a la Information Lens, Action Technologies Coordinator, etc). Seems like this sort of software would be very useful for using computers in an educational environment, as well as in the office. Higher education is already tending towards turning learning into an individual endeavor... I hope that the NeXT's vision of the future of higher education is not one in which we all plug into our little "smart TV's" and begin clacking away, oblivious to the world and to other people. -- Niels Mayer.
david@ms.uky.edu (David Herron -- One of the vertebrae) (10/25/88)
In article <2463@hplabsz.HPL.HP.COM> mayer@hplabsz.UUCP (Niels Mayer) writes: > >Could someone who has used the NeXT mailer comment on it's >functionality. While I haven't used their mailer I think I'm qualified to make some educated guesses :-) Voice/picture type stuff is usually in some binary format. There's no reason to suspect that this would be any different here. BUT, in order for it to be part of e-mail it has to be in printable ASCII. No funny characters, and some limits on what you can assume about tabs and/or end-of-line markers. Now, where to store it? It'll have to be part of the message somewhere. Putting it down in the body would be "inappropriate" because the system isn't really supposed to interpret things in the body, and also this is more an "envelope" type thing. A better place (er.. more appropriate) would be in the header somewhere. BUT, I can see a problem that this'd lead to large headers ... and I won't gaurantee that all systems or user-agents will be able to handle a header that's on the order of 4 or 5 Kbytes. (That's a bit beyond the norm for headers). What *I* am curious about is which mailer is at the transport level? -- <-- David Herron; an MMDF guy <david@ms.uky.edu> <-- ska: David le casse\*' {rutgers,uunet}!ukma!david, david@UKMA.BITNET <-- <-- Controlled anarchy -- the essence of the net.
guy@auspex.UUCP (Guy Harris) (10/27/88)
>>I've heard rumours that there are some connections between the Andrew >>project at CMU and NeXT. In particular, that NextStep is layered ontop >>of something akin to the andrew toolkit. > >NextStep is not based on Andrew. For one thing, it uses NFS as its basic >file-sharing mechanism. It IS based on the CMU operating system Mach. >Being only passingly familiar with Andrew, I can't comment on whether >or not they have duplicated Andrew features. He didn't say "Andrew", he said "the Andrew toolkit". Andrew includes a number of components, such as the Andrew file system, the Andrew window system server, and the Andrew window system toolkit. NeXTStEP, being more-or-less a window system toolkit (as opposed to an OS kernel or a file system), could well be based on the Andrew window system toolkit, even if the NeXT machine uses NFS rather than the Andrew file system.
wrs@Apple.COM (Walter Smith) (10/27/88)
In article <310@auspex.UUCP> guy@auspex.UUCP (Guy Harris) writes: >>>I've heard rumours ... that NextStep is layered ontop >>>of something akin to the andrew toolkit. >> >NeXTStEP, being more-or-less a window system toolkit (as opposed to an >OS kernel or a file system), could well be based on the Andrew window >system toolkit, even if the NeXT machine uses NFS rather than the Andrew >file system. Having used Andrew for a few years, and having seen lots of screen dumps from NeXT, I can say with great confidence that nExTsTeP has nothing at all to do with the Andrew toolkit. I believe it just on the basis of NeXtsTEp's appearance and alleged behavior, but if you want solid technical reasons, there are at least two: 1) ATK is based on CMU's own object-oriented C preprocessor, not Objective-C, and it wouldn't be easy to convert. 2) Steve Jobs would probably throw up if he used any application written with the Andrew toolkit. To avoid completely alienating myself from all my (former?) friends at CMU who may be working on ATK, I should mention that the current version was designed by a large group of very smart researchers who apparently like working quite independently, that making a truly smooth user interface on a multitasking machine is a very difficult task, and that ATK is in many ways a pioneer in the field of user interface toolkits. But I don't think any of that would change Steve's mind... - Walt -- Walter Smith wrs@apple.com, apple!wrs Apple Computer, Inc. (408) 974-5892 My corporation disavows any knowledge of my activities on the network.
mark@lakesys.UUCP (Mark Storin) (10/27/88)
Forgive my ignorance but, could someone summarize what the "Andrew Project" is? I have a friend is supposed to be getting access to an Andrew machine(?) and I'm curious as to what it's all about. -- Mark A. Storin Lake Systems, Milw., WI UUCP: uwvax!uwmcsd1!lakesys!mark || uunet!marque!lakesys!mark
bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (10/28/88)
Someone with the Andrew Toolkit (ATK) development group at CMU sent a note to info-andrew last week. Alas, I didn't save it, so I may be foggy on the details (not many of which will you find here, anyway). He described the directions ATK will be taking in the future, particularly with reference to X11R3. He also mentioned that a NeXT Step implementation of ATK (ATK layered on NeXT's window system, the way it is now layered on X11) was in the works. -=- Zippy sez, --Bob I always have fun because I'm out of my mind!!!
bader+@andrew.cmu.edu (Miles Bader) (10/29/88)
bob@allosaur.cis.ohio-state.edu (Bob Sutterfield) writes: > Someone with the Andrew Toolkit (ATK) development group at CMU sent a > note to info-andrew last week. Alas, I didn't save it, so I may be > foggy on the details (not many of which will you find here, anyway). > He described the directions ATK will be taking in the future, > particularly with reference to X11R3. He also mentioned that a NeXT > Step implementation of ATK (ATK layered on NeXT's window system, the > way it is now layered on X11) was in the works. It's actually a bigger project than that, as the port will use objective C (Barf, gag, "NO, IT'S NOT UGLY!") instead of our home-grown preprocessor. The port will probably use a lot of next's toolkit classes (in the interest of consistency with next software) instead of the low-level classes atk itself provided for X. Apparently, the next window system is much less seperable from their toolkit than X, anyway. The port WILL probably be data-stream compatible with existing versions of ATK (so you can exchange structured documents with them). -Miles [Take this message with a grain of salt; I'm observing all of this from the sidelines].
SCHAFER@RICE.BITNET (Richard A. Schafer) (10/31/88)
In article <10426@s.ms.uky.edu>, david@ms.uky.edu (David Herron -- One of the vertebrae) says: > >In article <2463@hplabsz.HPL.HP.COM> mayer@hplabsz.UUCP (Niels Mayer) writes: >> >>Could someone who has used the NeXT mailer comment on it's >>functionality. > > >While I haven't used their mailer I think I'm qualified to make >some educated guesses :-) > >Voice/picture type stuff is usually in some binary format. There's >no reason to suspect that this would be any different here. BUT, in order >for it to be part of e-mail it has to be in printable ASCII. No >funny characters, and some limits on what you can assume about tabs >and/or end-of-line markers. > >Now, where to store it? It'll have to be part of the message somewhere. >Putting it down in the body would be "inappropriate" because the system >isn't really supposed to interpret things in the body, and also this >is more an "envelope" type thing. A better place (er.. more appropriate) >would be in the header somewhere. BUT, I can see a problem that this'd >lead to large headers ... and I won't gaurantee that all systems or >user-agents will be able to handle a header that's on the order of >4 or 5 Kbytes. (That's a bit beyond the norm for headers). > >What *I* am curious about is which mailer is at the transport level? >-- ><-- David Herron; an MMDF guy <david@ms.uky.edu> ><-- ska: David le casse\*' {rutgers,uunet}!ukma!david, david@UKMA.BITNET ><-- ><-- Controlled anarchy -- the essence of the net. > I talked to the NeXT people about this at Educom'88, where they were demonstrating the system. The NeXT mail is supposedly "the standard Unix mail", and the voice extras are in fact transported in the body of the mail (at the end) as printable ascii characters, with (I believe) a piece of human readable text saying that the following is voice data. The NeXT modified code would read it and interpret the voice data. Richard Schafer #! rnews 450 Relay-Version: Version 1.7 PSU-NETNEWS 5/20/88; site PSUVM.BITNET Posting-Ve