plouff@nac.dec.com (Wes Plouff) (05/18/88)
DMouse 1.03 is really nice, but it has a couple of annoying bugs or
misfeatures. Here's what happens when it runs on my system, an A1000
with 512K, 2 drives, v1.2 operating system and no other software
enhancements or hacks. These are listed in order from minor bugs to
possible funny interactions.
1. At bootup, screen blanks after random time shorter than 5 minutes.
After that first time, screen blanking works normally.
2. At bootup, DMouse does not respond to Left-Amiga-esc comman until
after first mouse movement or click.
3. The s/startup-sequence lines
if EXISTS sys:system
path sys:system add
endif
etc. no longer add the 'system' and 'utilities' directories to the
file search path. (Possibly misunderstanding of CLI on my part.)
4. With click-to-front enabled, start Diskman v2.0. Click on gadget or
filename. The title bar disappears. Click right mouse button.
Title bar reappears. (Is the Diskman "title bar" actually a
separate window?)
Even with these flaws, DMouse has displaced PopCLI from my boot disk and
taken away all temptation to use MACH.
--
Wes Plouff plouff%nac.dec@decwrl.dec.com
Digital Equipment Corp, Littleton, Mass. or ...!decwrl!nac!plouffpapa@pollux.usc.edu (Marco Papa) (05/18/88)
In article <8805171931.AA07452@decwrl.dec.com> plouff@nac.dec.com (Wes Plouff) writes: |DMouse 1.03 is really nice, but it has a couple of annoying bugs or |misfeatures. |Even with these flaws, DMouse has displaced PopCLI from my boot disk and |taken away all temptation to use MACH. I LOVE DMOUSE! The only thing I am missing (and I know this is a sore point with Matt) is a "popup clock/timer". How can we convince Matt to add the clock to DMOUSE? Food? Money? :-) One reason I need a "popup" clock/timer if for upload/downloads. If Dmouse had it, I would even think about shipping it with every copy of A-Talk III. -- Marco Papa 'Doc' -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= uucp:...!pollux!papa BIX:papa ARPAnet:pollux!papa@oberon.usc.edu "There's Alpha, Beta, Gamma and Diga!" -- Leo Schwab [quoting Rick Unland] -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
doug-merritt@cup.portal.com (05/19/88)
Marco Papa writes: >I LOVE DMOUSE! The only thing I am missing (and I know this is a sore >point with Matt) is a "popup clock/timer". How can we convince Matt to >add the clock to DMOUSE? Food? Money? :-) I don't think this is appropriate...I asked Matt about adding a really trivial feature (making ^H work in string requesters), and he pointed out to me that the scope of DMOUSE is strictly limited. You can't add every cute/useful little feature in the world to something; that process gets out of hand quickly. (This is my view, I'm not quoting him directly). What you *really* want is a front-end interface language for Commodities Exchange. CX *is* the Perfect and General Method of handling any and all input handlers. Adding a standalone popup clock handler is easy, adding one to dmouse is easy, but the *right* place to put it is in a connection to CX. Jimm did a fantastic job on CX; anyone who hasn't taken the time to read the documentation on it yet is really missing out...it is an absolutely inspiring design. See Fish disk 87. My suggestion about a script language for CX isn't essential, it's just that it would then become trivial to add such features to CX. Actually, it's already *almost* trivial if you know how CX works. This would just take it that one last step. Doug --- Doug Merritt ucbvax!sun.com!cup.portal.com!doug-merritt or ucbvax!eris!doug (doug@eris.berkeley.edu) or ucbvax!unisoft!certes!doug
cjp@antique.UUCP (Charles Poirier) (05/19/88)
In article <9163@oberon.USC.EDU> papa@pollux.usc.edu (Marco Papa) writes: >I LOVE DMOUSE! The only thing I am missing (and I know this is a sore >point with Matt) is a "popup clock/timer". How can we convince Matt to >add the clock to DMOUSE? Food? Money? :-) >One reason I need a "popup" clock/timer if for upload/downloads. If Dmouse >had it, I would even think about shipping it with every copy of A-Talk III. Mach (latest is version 2.1) has a compile-time option for generating "Machclk" which has a clock. The clock can be toggled on or off, has modes for either auto pop-to-front or not, and a special mode for cost/minute so you can rate the cost of the current session on your favorite dial-up data service. Mach also has popcli, blanker, sunmouse, clicktofront (one or two click options), adjustable mouse speed, screen-to-front rA-M cycler (but sadly no window cycler), beeper, hot funkeys, one-hand mode, whatever else I may have forgotten, and features are all toggleable from the command line or while running. And it DOES include source. All in all, a very nice package. (I have no connection, am just a satisfied user of Mach.) -- Charles Poirier (decvax,ihnp4,attmail)!vax135!cjp "Docking complete... Docking complete... Docking complete..."
phil@titan.rice.edu (William LeFebvre) (05/19/88)
In article <8805171931.AA07452@decwrl.dec.com> plouff@nac.dec.com (Wes Plouff) writes: >DMouse 1.03 is really nice Agreed! I love having mouse acceleration. >3. The s/startup-sequence lines > if EXISTS sys:system > path sys:system add > endif > etc. no longer add the 'system' and 'utilities' directories to the > file search path. (Possibly misunderstanding of CLI on my part.) Hey! I've experienced the same phenomenon. Any CLIs I pop up have only one thing on their path: "sys:c". Maybe dmouse isn't starting it up in such a way that the path gets inherited correctly. I haven't tried to figure out just what the problem is: I first noticed it late last night as I was tired. Maybe this means it's time to switch over to a shell. William LeFebvre Department of Computer Science Rice University <phil@Rice.edu>
papa@pollux.usc.edu (Marco Papa) (05/20/88)
In article <2254@antique.UUCP| vax135!cjp (Charles Poirier) writes: |In article <9163@oberon.USC.EDU| papa@pollux.usc.edu (Marco Papa) writes: ||I LOVE DMOUSE! The only thing I am missing (and I know this is a sore ||point with Matt) is a "popup clock/timer". How can we convince Matt to ||add the clock to DMOUSE? Food? Money? :-) ||One reason I need a "popup" clock/timer if for upload/downloads. If Dmouse ||had it, I would even think about shipping it with every copy of A-Talk III. | |Mach (latest is version 2.1) has a compile-time option for generating |"Machclk" which has a clock. The clock can be toggled on or off, has |modes for either auto pop-to-front or not, and a special mode for |cost/minute so you can rate the cost of the current session on your |favorite dial-up data service. Mach also has popcli, blanker, |sunmouse, clicktofront (one or two click options), adjustable mouse |speed, screen-to-front rA-M cycler (but sadly no window cycler), |beeper, hot funkeys, one-hand mode, whatever else I may have forgotten, |and features are all toggleable from the command line or while running. |And it DOES include source. All in all, a very nice package. |(I have no connection, am just a satisfied user of Mach.) Sorry, but I have to disagree. I got Mach (1.6?) [thanks for Bob and the other guy that mailed it to me], and I tried it and it just crashes and misbehaves TOO much. It locks up too many times, and it misbehaves on GADGETS. So I find it unusable. I guess I'm sold on Dmouse and will be waiting for a popup/clock/timer [ala sidekick] to surface. The current crop of clocks [including the one in mach] just don't cut it. -- Marco Papa 'Doc' -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= uucp:...!pollux!papa BIX:papa ARPAnet:pollux!papa@oberon.usc.edu "There's Alpha, Beta, Gamma and Diga!" -- Leo Schwab [quoting Rick Unland] -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
ejkst@cisunx.UUCP (Eric J. Kennedy) (05/21/88)
In article <9206@oberon.USC.EDU> papa@pollux.usc.edu (Marco Papa) writes: >on GADGETS. So I find it unusable. I guess I'm sold on Dmouse and >will be waiting for a popup/clock/timer [ala sidekick] to surface. >The current crop of clocks [including the one in mach] just don't >cut it. I guess I don't really know what you are looking for in a clock, but I've been using a tiny little clock called wclock for a while now. It's only about 640 bytes on disk. The window is only big enough for the time of day and doesn't do a timer, date, or anything else. what I like about it is that it pops to the front every time it changes the time, so it always comes to the front even after you rearrange windows. I have no idea who wrote it or even where I got it. I don't ever remember seeing a doc or readme file for it. I just like it, that's all. >-- Marco Papa 'Doc' -- ------------ Eric Kennedy ejkst@cisunx.UUCP
peter@sugar.UUCP (Peter da Silva) (05/22/88)
In article <5593@cup.portal.com>, doug-merritt@cup.portal.com writes: > My suggestion about a script language for CX isn't essential, it's just > that it would then become trivial to add such features to CX. Actually, > it's already *almost* trivial if you know how CX works. This would just > take it that one last step. AREXX is a decent scripting language. It's not well suited to handling things other than text strings directly, but you can put in a level of abstraction to hide the stuff. I wouldn't want to try to pass input messages through it, but as a tool to build trees it'd work fine. Use integers or other tokens to tell REXX about the nodes... -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These may be the official opinions of Hackercorp.
doug-merritt@cup.portal.com (05/23/88)
I wrote about creating a script language for Commodities Exchange. Peter da Silva replied: >AREXX is a decent scripting language. It's not well suited to handling >things other than text strings directly, but you can put in a level of >abstraction to hide the stuff. I've been staying out of the AREXX discussion so far, mainly because it seems clear to me that there are good arguments on both sides. But let's not take it too far...your enthusiasm has taken you to a point where you could say the same thing about *anything*, like if I ask for a startup script language for VT100 (it already has one, btw) then you say, "oh, 'script language', that's my cue...USE AREXX!!!" Don't pound square pegs into round holes! No matter how wonderful AREXX is, the simple fact of the matter is that very few people have it, and at $50 a shot, it's going to stay that way. If you insist that everything use AREXX, then we'll have to wait until everyone does have it to implement anything new. More logical would be to create a CX script language now. Later, when 1.4 includes AREXX as part of the standard distribution, *then* we can use it with CX (if you still think it's appropriate, which I have strong doubts about). Until then we need interim solutions. Hmmm...I didn't use any arguments about technical merit. Ok, consider this...what I was talking about for CX doesn't need *any* of the features of AREXX. It doesn't need pattern matching, it doesn't need real smart IPC control, it doesn't need control structures. All you need is something real simple, like a file specifying what program to start up when some hotkey is pressed, and translations specifying that hotkey A should be translated to hotkey B, etc. You could write AREXX programs to do this, sure, but it'd be a lot harder than it would be with a simple script language. And it would be overkill to boot. CX already supports all of the *functionality* you need; it's just that currently you need to tell it what you want via a C program. I'm proposing a more human friendly interface, that even non-programmers could use, that's all. It's not like you to have knee jerk reactions, Peter. Usually you think things out very clearly. Oh well...I've posted much dumber things myself, so I shouldn't criticize...With all the AREXX debate going on, it must be easy to get jumpy. Doug -- Doug Merritt ucbvax!sun.com!cup.portal.com!doug-merritt or ucbvax!eris!doug (doug@eris.berkeley.edu) or ucbvax!unisoft!certes!doug
lphillips@lpami.van-bc.UUCP (Larry Phillips) (05/27/88)
In <5768@cup.portal.com>, doug-merritt@cup.portal.com writes: >Don't pound square pegs into round holes! No matter how wonderful >AREXX is, the simple fact of the matter is that very few people have it, >and at $50 a shot, it's going to stay that way. If you insist that >everything use AREXX, then we'll have to wait until everyone does >have it to implement anything new. > >More logical would be to create a CX script language now. Later, when >1.4 includes AREXX as part of the standard distribution, *then* we >can use it with CX (if you still think it's appropriate, which I have >strong doubts about). Until then we need interim solutions. The only problem with what you say is that I, and probably most people, have seen very little of CX. All I ever saw was a 'proposal' sort of thing that described the Commodities Exchange, without going into a lot of detail, and without any binaries/source/whatever. Yes, ARexx is a $50 package, and yes, not everyone has it, but to many people, right now, it is a lot more available than CX, and is getting ever increasing support from developers. Looking at it another way, would you consider $50 to be an outrageous price for a fine quality tool? >Hmmm...I didn't use any arguments about technical merit. Ok, consider >this...what I was talking about for CX doesn't need *any* of the features >of AREXX. It doesn't need pattern matching, it doesn't need real smart >IPC control, it doesn't need control structures. Right. They sound to me like two completely different things >All you need is something real simple, like a file specifying what >program to start up when some hotkey is pressed, and translations >specifying that hotkey A should be translated to hotkey B, etc. You >could write AREXX programs to do this, sure, but it'd be a lot >harder than it would be with a simple script language. And it would be >overkill to boot. CX already supports all of the *functionality* you >need; it's just that currently you need to tell it what you want via >a C program. I'm proposing a more human friendly interface, that even >non-programmers could use, that's all. Perhaps I am remembering the one article I read on CX badly, but I didn't think it was any more than an arbiter of hotkey combinations? Is there more? If so, please tell me more about it. >It's not like you to have knee jerk reactions, Peter. Usually you think >things out very clearly. Oh well...I've posted much dumber things myself, >so I shouldn't criticize...With all the AREXX debate going on, it must be >easy to get jumpy. Are you sure it was a knee jerk reaction? Sounds to me like Peter got caught up in the simplicity and power of ARexx, and passed on some of that enthusiasm. -larry -- If all the MSDos machines were laid end to end, they still wouldn't be as fun as a single Amiga. +----------------------------------------------------------------+ | // Larry Phillips | | \X/ {ihnp4!alberta!ubc-vision,uunet}!van-bc!lpami!lphillips | | COMPUSERVE: 76703,4322 | +----------------------------------------------------------------+
doug-merritt@cup.portal.com (05/30/88)
Larry Phillips writes: >[I] have seen very little of CX. All I ever saw was a 'proposal' sort of thing >that described the Commodities Exchange, without going into a lot of >detail, and without any binaries/source/whatever. CX was distributed on Fish disk 87, both documentation and binaries. It was also locally distributed by Jimm at a BADGE meeting. >Yes, ARexx is a $50 >package, and yes, not everyone has it, but to many people, right now, it is >a lot more available than CX, and is getting ever increasing support from >developers. Looking at it another way, would you consider $50 to be an >outrageous price for a fine quality tool? That's not the point at all. For one thing, *both* are available, and CX is free. For another thing, it is wholly intended to be used to do input handlers in a clean way. Therefore I was saying that if you want to do input handlers, CX is the right way. Using AREXX for an input handler just doesn't make any sense at all. > Perhaps I am remembering the one article I read on CX badly, but I didn't >think it was any more than an arbiter of hotkey combinations? Is there >more? If so, please tell me more about it. The whole problem is that it isn't well understood. Half the point of my comments was to remind people that it existed, and to generate some interest in programmers using it. It's *beautifully* designed and executed, and it's the perfect tool *for programmers* to use for input handlers. I'm not prepared to define its scope, but it's quite powerful. Bringing up AREXX in this context is about as irrelevent as if I started talking about using Basic...there's simply no connection whatsoever. >Sounds to me like Peter got caught up in the simplicity and power of >ARexx, and passed on some of that enthusiasm. Understandable, yet illogical in this context. One thing I talked about may be especially unclear: the idea of creating a "script language" for CX. I didn't think this out in much detail before saying something, so my comments may have been misleading. The idea was just to make CX available to more casual use than the current audience of systems programmers. Upon reflection, it seems to me that this "script language" might actually be a C-code program generator (although in a trivial sense). Anyway, to follow up on this point, it'd be a good idea to check out CX on Fish 87 first. I need to re-read the docs myself. Doug -- Doug Merritt ucbvax!sun.com!cup.portal.com!doug-merritt or ucbvax!eris!doug (doug@eris.berkeley.edu) or ucbvax!unisoft!certes!doug
peter@sugar.UUCP (Peter da Silva) (06/01/88)
In article ... lphillips@lpami.van-bc.UUCP (Larry Phillips) writes: > The only problem with what you say is that I, and probably most people, > have seen very little of CX. All I ever saw was a 'proposal' sort of thing > that described the Commodities Exchange, without going into a lot of > detail, and without any binaries/source/whatever. It's on a fish disk. What more do you want? > Perhaps I am remembering the one article I read on CX badly, but I didn't > think it was any more than an arbiter of hotkey combinations? Is there > more? If so, please tell me more about it. CX is a general arbiter of the Amiga input stream. That includes hotkeys, macrose, redirected mouse movements, etc... > Are you sure it was a knee jerk reaction? Sounds to me like Peter got > caught up in the simplicity and power of ARexx, and passed on some of that > enthusiasm. Power, yes. Sort of. I wish it was a little less text oriented. Simplicity, no. It's not a simple language. Its main avantage is that it already exists. Actually, it's about what I expected. -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These may be the official opinions of Hackercorp.
peter@sugar.UUCP (Peter da Silva) (06/01/88)
In article <5768@cup.portal.com>, doug-merritt@cup.portal.com writes: > I wrote about creating a script language for Commodities Exchange. > Peter da Silva replied: > >AREXX is a decent scripting language. It's not well suited to handling > >things other than text strings directly, but you can put in a level of > >abstraction to hide the stuff. > Hmmm...I didn't use any arguments about technical merit. Ok, consider > this...what I was talking about for CX doesn't need *any* of the features > of AREXX. It doesn't need pattern matching, it doesn't need real smart > IPC control, it doesn't need control structures. AREXX doesn't have real smart IPC control. It has the ability to send a text string to a named port, and get a text string in response. But that's quibbling. It may well be overkill. > harder than it would be with a simple script language. And it would be > overkill to boot. CX already supports all of the *functionality* you > need; it's just that currently you need to tell it what you want via > a C program. I'm proposing a more human friendly interface, that even > non-programmers could use, that's all. Now a general-purpose scripting package is also a good idea. I've been thinking of Xlisp, but it's pretty big (over 64K). A small forth-like language would be ideal. A tiny lisp would be better. Matt Dillon's DME/DTERM stuff is OK, though the lack of variables in DTERM is a big pain in the neck. Perhaps encapsulating these routines would be a good idea. Or maybe ripping the interpreter stuff out of Micro-Emacs. > It's not like you to have knee jerk reactions, Peter. Usually you think > things out very clearly. Thanks for saying so. I try. -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These may be the official opinions of Hackercorp.
peter@sugar.UUCP (Peter da Silva) (06/01/88)
I have a little clock you might like. It's got a little RPN calculator hidden in it... you select it and hit the space bar. The date vanishes and you get a little 32-bit integer only calculator display that handles hex, decimal, and octal. I was tired of all those stupid calculators that took up all that screen space with a redundant keyboard. I'd post it to sources, but I think they have enough on their hands right now. If enough people want a copy I'll do that, otherwise I'll just mail it. -- -- Peter da Silva `-_-' ...!hoptoad!academ!uhnix1!sugar!peter -- "Have you hugged your U wolf today?" ...!bellcore!tness1!sugar!peter -- Disclaimer: These may be the official opinions of Hackercorp.