I have a few more questions on ILISP and Emacs.
1) ILISP User Guide
First of all, the ILISP User Guide (ilisp.pdf) appears to be compiled
from a fasimile instead of text. I don't mind the whole thing looking
like a low-res fax, but this is very bad news if you want to do any
searches within the document. You won't find anything :-(
Is there any chance to get the document in ASCII or RTF ?
2) HyperSpec Lookup
How is one supposed to invoke the hyperspec lookup?
Well, <C-z> doesn't work, so I can't do <C-z> H (hyperspec-lookup). In
fact there is little I can do at all with most keyboard shortcuts not
working. I tried the commandline version of Emacs, which is even
worse, can't even quit the program because keyboard shortcuts don't
work at all - none. :-(
Anyway, leaving the keyboard mess aside for the moment, I try to bring
up the hyperspec from the menu (ILisp -> Documentation -> hyperspec -
apropos). This doesn't work either. Instead it generates an error
message "Wrong type argument: commandp, hyperspec-lookup".
I have downloaded HyperSpec version 6 and placed it in
/Developer/Documentation/Lisp/HyperSpec
Accordingly, I inserted the following in my .ilisp file ...
==============================================================================
;;; Configuration of Erik Naggum's HyperSpec access package.
;; If you have a local copy of the HyperSpec, set its path here.
(setq common-lisp-hyperspec-root
"file:/Developer/Documentation/Lisp/HyperSpec/")
(setq hyperspec-root-url "/Developer/Documentation/Lisp/HyperSpec/"
hyperspec-symbol-table
"/Developer/Documentation/Lisp/HyperSpec/Data/Map_Sym.txt")
;;; Sample load hook
(add-hook 'ilisp-load-hook
'(lambda ()
;; Change default key prefix to C-c
(setq ilisp-*prefix* "\C-c")
;; Set a keybinding for the COMMON-LISP-HYPERSPEC command
(defkey-ilisp "\C-cH" 'hyperspec-lookup-lisp)
;; Make sure that you don't keep popping up the 'inferior
;; Lisp' buffer window when this is already visible in
;; another frame. Actually this variable has more impact
;; than that. Watch out.
(setq pop-up-frames nil)
(message "Running ilisp-load-hook")
;; Define LispMachine-like key bindings, too.
; (ilisp-lispm-bindings)
Sample initialization hook.
;; Set the inferior Lisp directory to the directory of
;; the buffer that spawned it on the first prompt.
(add-hook 'ilisp-init-hook
'(lambda ()
(default-directory-lisp ilisp-last-buffer)))
))
==============================================================================
before the line that says
(setq lisp-mode-hook '(lambda () (require 'ilisp)))
3) Access to the clipboard
How can I copy-paste something from outside of Emacs into Emacs?
With any Mac applications that run under Aqua, I can cut or copy
something in one application and then paste it in another application.
Well, Emacs (though being an Aqua application) doesn't seem to know
anything about the clipboard. If I go to the Edit menu, "Paste" is not
available although I have just copied something in the clipboard.
Also, on the Mac, it is absolutely normal to just drag a snippet of
text with the mouse from one application and drop it into another
application, without having to do any copy and paste. Again, Emacs
(although an Aqua application) doesn't seem to be a good team player.
4) Fonts
How does one change the font Emacs is using?
tia
rgds
bk
·········@yahoo.co.uk (BK) wrote in message news:<····························@posting.google.com>...
> 1) ILISP User Guide
>
> First of all, the ILISP User Guide (ilisp.pdf) appears to be compiled
> from a fasimile instead of text. I don't mind the whole thing looking
If you refer to the PDF version of the manual that comes with the
ILISP documentation snapshot, this is probably due to the fact that I
generated it with an old version of ghostscript. I am about to do a
major hardware/software upgrade, but that's what I can do now.
> Is there any chance to get the document in ASCII or RTF ?
What about the PostScript version of the manual that also comes with
the documentation snapshot? Are you able to open and read it? If
Texinfo is available for Mac OS X, you can also try to build the
manual from source.
I can't provide more help now, sorry.
Paolo
·········@yahoo.co.uk (BK) writes:
> I have a few more questions on ILISP and Emacs.
>
> 1) ILISP User Guide
The source for the ILISP package contains (in theory) what you need to
generate the .info, .pdf, .ps, etc files. I haven't tried building
ILISP on my Mac (OS X 10.2.5). However, I built the pdf file on my
Debian Linux box and it looks ok under Preview.
http://www.david-steuber.com/pdf/ilisp.pdf
Hope that helps.
> 2) HyperSpec Lookup
>
> How is one supposed to invoke the hyperspec lookup?
>
> Well, <C-z> doesn't work, so I can't do <C-z> H (hyperspec-lookup). In
I am hacking at my Mac in spurts to get various things working. Being
a Unix guy, I have gotten Fink (off of sourceforge) and X11 from
Apple. C-h acts as a backspace in my xterm (I thought I had that
fixed :-/). However, the Terminal program has C-h work to let you get
at help (running emacs 21 inside Terminal). Sadly, C-z still throws
emacs into the background. I haven't got that sorted. I think the
thing to do is remap C-z to something else.
> 3) Access to the clipboard
The Terminal program works with Command-C and Command-V. Apple's
xterm (from X11) works pasting the apple clipboard when you press
mouse-2. There are preferences to allow you to emulate a three button
mouse for X11. Actually, Command-C should work too. It is in the
Edit menu.
> 4) Fonts
This I have not tried. I am not using the carbon emacs. Emacs 21 is
supposed to support different fonts though.
David Steuber <·············@verizon.net> writes:
> ·········@yahoo.co.uk (BK) writes:
>
> > 4) Fonts
>
> This I have not tried. I am not using the carbon emacs. Emacs 21 is
> supposed to support different fonts though.
I'm pretty sure that under all platforms, shift+left-click pulls up a
font menu. For more Emacs help, you should ask on gnu.emacs.help.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
David Steuber <·············@verizon.net> wrote ...
> I built the pdf file on my
> Debian Linux box and it looks ok under Preview.
>
> http://www.david-steuber.com/pdf/ilisp.pdf
>
> Hope that helps.
great. thanks. Paolo, you might want to pick this up, too ;-)
> > 2) HyperSpec Lookup
> >
> > How is one supposed to invoke the hyperspec lookup?
> >
> > Well, <C-z> doesn't work, so I can't do <C-z> H (hyperspec-lookup). In
>
> I am hacking at my Mac in spurts to get various things working. Being
> a Unix guy, I have gotten Fink (off of sourceforge) and X11 from
> Apple. C-h acts as a backspace in my xterm (I thought I had that
> fixed :-/). However, the Terminal program has C-h work to let you get
> at help (running emacs 21 inside Terminal).
Nope. My /usr/bin/emacs does not react to *any* keyboard shortcuts,
nothing, zero, nil !!! I can't even quit the damn thing.
> Sadly, C-z still throws emacs into the background. I haven't got that
> sorted. I think the thing to do is remap C-z to something else.
I am not going to waste my time on this excuse for an editor, I won't
touch it, not even with a pair of pliers. It's either Aqua or nothing.
> > 3) Access to the clipboard
>
> The Terminal program works with Command-C and Command-V. Apple's
> xterm (from X11) works pasting the apple clipboard when you press
> mouse-2. There are preferences to allow you to emulate a three button
> mouse for X11. Actually, Command-C should work too. It is in the
> Edit menu.
I don't want to have to install X11 only to run an editor every time I
am on another desktop that doesn't have X11 installed.
See, I am not really interested in Emacs, per se. In fact, Emacs is so
horrible when it comes to user friendliness that I should expect to
get paid for using it, but unlike many programmers here, I am not
getting paid for this. For me it is supposed to be fun.
What I *really* want to do is play with Lisp. So far Emacs/ILISP
looked like it would add value by *enabling* me to do what I really
want to do. Now, it looks more and more as if Emacs is a *road block*
that stops me from doing Lisp.
Perhaps, I shold cut my losses and bypass Emacs altogether. I have
downloaded a really nice editor which is written in CL, called Alpaca
http://sourceforge.net/project/showfiles.php?group_id=79717
Emacs developers should take a look at this and get a clue how to make
something actually user friendly in the 21st century.
Alpaca is built with OpenMCL using Cocoa and it supports drag and
drop. It has an option to open an OpenMCL listener and comes with all
sources. It even supports Emacs keybindings and of course being
written in Lisp it is extensible with Lisp just like Emacs is, even
better it uses CL.
It is probably not much more effort to add the HyperSpec and paren
matching to either Alpaca or the OpenMCL "IDE" than it is to try and
fix Emacs. At present I lack the skills to do either. But at least
trying to do the former will get me exposure to CL, which is what I am
after.
Anyway, if someone knows how to get the Hyperspec working with Emacs
and ILISP, I am happy to try suggestions in order to be able to
incorporate it into the step-by-step guide I have promised - I keep my
promises.
anyway, thanks for comiserating with me
rgds
bk
·········@yahoo.co.uk (BK) writes:
> Alpaca is built with OpenMCL using Cocoa and it supports drag and
> drop. It has an option to open an OpenMCL listener and comes with all
> sources. It even supports Emacs keybindings and of course being
> written in Lisp it is extensible with Lisp just like Emacs is, even
> better it uses CL.
I'm sorry I couldn't be more help. I'm just getting started also.
I'll have to check out this Aplpaca thing if I can't set up a
duplicate dev environment on my Mac. I'm trying to be cross-platform.
David Steuber <·············@verizon.net> wrote ...
> I'm sorry I couldn't be more help.
That's alright.
> I'm just getting started also.
> I'll have to check out this Aplpaca thing if I can't set up a
> duplicate dev environment on my Mac. I'm trying to be cross-platform.
Alpaca is really nice. Also I find the source code to have good
educational value. Well, at least for me ;-) However, Alpaca uses the
Cocoa Text Application framework, so it is not entirely Lisp aware,
but the author says he wants to make it a good Lisp editor, so let's
see how this evolves.
Have you been able to find out why most of the Emacs keyboard
shortcuts are deaf?
I tried to get some help on this in the emacs help forum, but that was
like smashing into a hornet's nest. It seems those Emacs folks are
religious fanatics. You are not supposed to mention anything that
doesn't work, or so it would seem ;-)
However, I would really like to complete a newbie proof how-to-set-up
Carbon Emacs and ILISP guide because I think that this is a stumbling
block.
How is Lisp going to win more friends if you have to fight with the
editor for a week?! That will scare away anyone but the most
determined. Perhaps the old-hands in the Lisp community like it that
way, I don't know. Perhaps the Lisp community is an elitist club and
they want nobody but the most determined to join them, I don't know.
I have run into a few things that suspiciously smelled this way. And I
have seen this before with the Mac crowd, when OSX came out. Suddenly
the Mac was popular amongst Unix folks and the Mac die hards didn't
like it. Their little elitist Mac world was threatened and they blamed
OSX for this. Amazing how people project themselves onto machines and
develop fetishes, isn't it?! :-)
Anyway, I am still determined to find out and document everything a
newbie must know to get Carbon Emacs and ILISP on OSX into a
reasonably working state so they can go straight to Lisp. So, if you
find out anything, please keep me in the loop.
thanks
rgds
bk
·········@yahoo.co.uk (BK) writes:
> I tried to get some help on this in the emacs help forum, but that was
> like smashing into a hornet's nest. It seems those Emacs folks are
> religious fanatics. You are not supposed to mention anything that
> doesn't work, or so it would seem ;-)
I think *you* were the hornets and the nest in this situation -- for
anyone curious, he titled the thread "Is Emacs on Aqua crippleware or
is it just broken?", and highlights of his rantings include: "See, I
give a damn about that excuse for an editor."
True to its usual form, gnu.emacs.help was remarkably calm and helpful
in response to his flamefest.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
···@fallingrocks.OCF.Berkeley.EDU (Thomas F. Burdick) wrote ...
> I think *you* were the hornets and the nest in this situation -- for
> anyone curious, he titled the thread "Is Emacs on Aqua crippleware or
> is it just broken?", and highlights of his rantings include: "See, I
> give a damn about that excuse for an editor."
Quoting out of order and context seems to be a popular sport these
days. You also conveniently forgot to mention that I apologised for
that heat-of-the-moment sentence which was the only one of its kind.
My original post was precise, fair and just and yes, I can get
provoked and react less nicely, just like any other human being.
You should have seen the Coke and Pepsi experiment, or perhaps you
have. It went like this. A Coca-Cola afficionado was give to taste two
glasses of Pepsi and asked to determine which one was the Pepsi.
Needless to say he guessed right. Same procedure for the Pepsi-Cola
afficionado accordingly with two glasses of Coke. He too guessed
right. Next they were asked to characterise the drink in the other
glass which they thought was the other, evil brand. Not surprisingly,
they had nothing good to say about it. When they were faced with the
truth they had a nervous break down, realising they had just smashed
their own religion into pieces.
<Warning! UK style sarcasm included below>
Any bet if you'd be able to recreate that experiment on the basis of
the behaviour of the Carbon Emacs I have sitting here (and this is the
least troublesome out four) you'd get to hear things directly from the
horse's mouth that I couldn't even begin to think of, let alone
express.
See, I can get together with vi folks and say "vi - what a horrible
and incomprehensible mess it is" and they'd agree with me, they'd
smile and they'd say "yeah, but we have come to really like it" and I
can smile back.
I can get together with sendmail folks and say "sendmail - what a
butt-ugly incomprehensible crawling monster hog" and they'd agree with
me, they'd smile and they'd say "yeah, but we have come to really like
it" and I can smile back.
I could list more examples of tools where its users and afficionados
know what's wrong and they can laugh about it. Only very few folks
make a religion out of their tools. Emacs, is apparently such a case.
Mind you, not my judgement, but that of a number of self declared
members to whom I have since spoken to.
Well, I should have realised that something was odd when I saw the
built-in virtual psychiatrist. That should have set off the alarm
bells. Now I see.
Hopefully I will never develop any fetish for neither hardware nor
software, but should I ever be so unfortunate to come close to that
state of mind, I pray to God that I will have enough reason left to
realise that it is time to take a break and not touch any computer for
at least a number of years.
But anyway, you want to have some fun at my expense? Go ahead, be my
guest, don't be shy. I shall take pride in being the one who can take
more beating than anybody else and still have a laugh while taking it
;-)
rgds
bk
I have no idea why you are having trouble with the meta and ctrl keys
in carbon emacs. I haven't figured out the deal with GNU emacs yet
;-). I imagine the Mac just twiddles different bits.
It so happens I am one of the Unix people who came over to OS X
because it had Unix under the hood. It is rather different in flavor
than Debian, but at least I can understand a chunk of it. And as much
as I enjoy the Aqua interface, I like being able to use a command
line.
Right now, I am on my Mac, but I have ssh -X over to my Debian box to
use GNU emacs in an X11 window. Mozilla just wasn't cutting it as a
news reader. Now I am using GNUS. The one oddity is that I have to
use the ESC key for meta. But I have to tap it. If I hold it too
long, it becomes ESC (would that by the hyper key?).
All this software mixing makes for some complex soup.
Anyway, I am reducing the number of nuts I have to crack at once. I
am going with GNU emcas/ILISP on Debian. CMUCL will be the Lisp
flavor I play with (I know that CLisp is available on both Debian and
Mac OS X). I will work on learning the environment, then Lisp.
After I am comfortable with Lisp (I plan to go through Graham's books
and do the more interesting excercises), I will see just how hard it
is to steal an idea back from Java. I am aware of CLIM and Garnet. I
want to see if I can develop my own cross platform GUI API along the
lines of how Swing/AWT works.
Heck, I would love to do my own CL implementation. Probably never
going to happen, but I can dream. It really comes down to how quickly
I can grok all this stuff. Certain forms are very obviouse to me
while others are like, "huh?".
My question would by, "why carbon?" I think cocoa would taste better
;-). CLX seems to be the root of evil on X11. Cocoa is the logical
API for OS X. And if it ever came to Windows, there is the GDI.
OpenGL has already been done (I think) for 3D. I do not know if there
is a CLOS flavor of OpenInventor.
One oddity in my Lisp journey is that I got into Linux for its free
Lisp implimentations. Then I got distracted. I hope I don't get
distracted again, because from all I have read, I think Lisp really is
a key to productive programming. Whatever roadbumps I have to deal
with initially are probably no worse than with C++ programming. I was
able to pick that up (on Windows).
There is a comfort zone one reaches where they can suddenly be
productive. The key is to not give up before that point.
David Steuber <·············@verizon.net> wrote ...
> I have no idea why you are having trouble with the meta and ctrl keys
> in carbon emacs.
Neither have I ;-) Mind you this is the least troublesome of four!
builds.
Somebody mentioned yet another build, which I am going to try out
later today.
> It so happens I am one of the Unix people who came over to OS X
> because it had Unix under the hood.
Same here, but I made the switch already a long time ago. I was coming
from VAX/VMS via Ultrix to the Mac running MachTen, which was a
commercial BSD that ran on top of and in parallel with the MacOS
(before OSX). In the process of using the Mac, I got a bit spoiled
over the years ;-)
> All this software mixing makes for some complex soup.
One of the reasons I'd like to set up an environment based on Aqua.
> I know that CLisp is available on both Debian and Mac OS X).
Interesting. Perhaps I should include the ILISP setup for that in my
mini-how-to then.
> I will work on learning the environment, then Lisp.
My aim is to make the hurdle to get to the Lisp part of it as low as
possible.
> After I am comfortable with Lisp (I plan to go through Graham's books
> and do the more interesting excercises), I will see just how hard it
> is to steal an idea back from Java. I am aware of CLIM and Garnet. I
> want to see if I can develop my own cross platform GUI API along the
> lines of how Swing/AWT works.
In another life, I once learned VAX-Lisp at university and every now
and then I tried to pick it up again going through tutorials and
books. This has given me a good grasp of the concepts but left me
without any practical skills ;-)
Therefore, this time, I'd like to try and jump into something
practical. The key is to find something that is real-world enough but
not too difficult to do. I was thinking about trying to play with the
Cocoa based example of the rudimentary IDE that comes with OpenMCL and
extend it with some UI stuff. Not necessarily something that adds
significant value, but something that is simple enough to do. I was
thinking perhaps adding a simple user preferences window to set
default window size, font and font size using the Cocoa Preferences
framework may be such a thing. Then again, that may already be too
ambitious, I guess I'll find out soon ;-)
> Heck, I would love to do my own CL implementation. Probably never
> going to happen, but I can dream. It really comes down to how quickly
> I can grok all this stuff. Certain forms are very obviouse to me
> while others are like, "huh?".
Oh well, Lisp is an incredibly rich language. On the outset it looks
like there is almost no syntax, everything's a list denoted by parens,
that's simple, right? ;-) But this is precisely the reason that makes
it so flexible and it would be surprising if that had not led to the
richness it has today after more than 40 years.
I am always puzzled when people say something like "Lisp would be OK
if it wasn't for the parens". Don't they realise that this is exactly
what makes the language so powerful? If it wasn't for the lists and
parens you'd either have to have some specialised syntax and semantics
to define new syntax, which is by no means an easy task to implement,
or you'd be limited to the constructs the original language designers
thought to be sufficient. Lisp is still around because you don't have
to abandon the language and create a new one everytime you run into a
limitation severe enough to make it seem worth the trouble to start
from scratch.
Before that background it seems to me that it is justifiable that it
takes a little longer to master the language. Then again, do you have
to *master* the language in order to be productive using it anyway?
If I learn a new natural language, it will take me years of practise
to get a native level proficiency, *BUT* I can pick up enough to
communicate in daily life situations within only a few weeks. And that
may be just be good enough. If I am going to Italy on holidays I don't
need to be proficient in Italian, it's perfectly sufficient to know
just enough to understand what the newspaper headlines say, to read a
menu and order a meal in a restaurant, to shop and to ask my way
around. That level I can reach within a few days if I am proficient in
at least one other Latin language. I won't be able to understand the
libretto when I visit the opera, but I can enjoy the music instead.
So, that's my attitude to Lisp. If I can get to the point where I can
use it comfortably as a scripting language, that'll be just fine. I
won't win any beauty contest nor any F1 race, but I don't care. If I
need to win a beauty contest, I can hire a model, if I need to win a
F1 race, I can hire a driver.
The trouble I have with Emacs right now is like I am trying to learn
Italian with a text book that is printed upside down, with the
characters mirrored and every other word is mangled. If you can get
used to that, people say it's a great text book, but I personally
would rather use a textbook, I don't have to get used to. However, I
started this mini-how-to, so I have to finish that.
> My question would by, "why carbon?" I think cocoa would taste better
> ;-). CLX seems to be the root of evil on X11. Cocoa is the logical
> API for OS X. And if it ever came to Windows, there is the GDI.
> OpenGL has already been done (I think) for 3D. I do not know if there
> is a CLOS flavor of OpenInventor.
For all practical purposes, Carbon should be seen as an aid to port
legacy apps to OSX. If you don't do legacy apps, take the time to
learn Cocoa. I have been using AppleScript Studio which is basically
an automated Cocoa wrapper generator for AppleScript. Cocoa is a very
powerful framework and it seems well designed and mature. I just can't
get warm with Objective-C, or any C for that matter. If I want a
static language with type checking and all that, I want something very
strict, something Algolish or Pascalish, or even stricter, with data
encapsulation like Modula or whatever the flavour of the day of that
line is today. But if I want something flexible, I want real
flexibility even at the expense that it totally overwhelms me,
something like Lisp. So, I hope there is a future for Cocoa+Lisp,
otherwise I may have to stick with AppleScript Studio.
> One oddity in my Lisp journey is that I got into Linux for its free
> Lisp implimentations. Then I got distracted.
That's the trouble with all things computer. You always get distracted
with some other stuff before you can do what you set out to do and
eventually you risk losing your return pointer altogether. That's what
I have come to appreciate about the Mac. It's not free of this danger,
but it is a lot closer to realising the idea of not letting technology
get in your way than anything else I have seen.
> I hope I don't get
> distracted again, because from all I have read, I think Lisp really is
> a key to productive programming. Whatever roadbumps I have to deal
> with initially are probably no worse than with C++ programming. I was
> able to pick that up (on Windows).
>
> There is a comfort zone one reaches where they can suddenly be
> productive. The key is to not give up before that point.
Certainly. I also think it is important to try to avoid getting
sidelined with things that may seem to be necessary to reach that
point but in reality they are just another unneccessary distraction.
rgds
bk
David Steuber <·············@verizon.net> wrote:
> I have no idea why you are having trouble with the meta and ctrl keys
> in carbon emacs. I haven't figured out the deal with GNU emacs yet
> ;-). I imagine the Mac just twiddles different bits.
Just a thought: are the control and meta available to the carbon emacs at
all?
Whenever there has been ctrl/meta troubles on my boxes it has always
turned out to be a case of the window manager hogging them...
Cheers,
-- Nikodemus
Nikodemus Siivola <········@kekkonen.cs.hut.fi> wrote ...
> Just a thought: are the control and meta available to the carbon emacs at
> all?
yes, some combinations work - some don't
I am now trying yet another build and see how that one goes ;-)
> Whenever there has been ctrl/meta troubles on my boxes it has always
> turned out to be a case of the window manager hogging them...
interesting. could it "hog" just some combinations but not others?
thanks
rgds
bk
BK <·········@yahoo.co.uk> wrote:
>> Whenever there has been ctrl/meta troubles on my boxes it has always
>> turned out to be a case of the window manager hogging them...
> interesting. could it "hog" just some combinations but not others?
I doubt it. But then the boxes I referred to weren't Macs... By the
way: have you actually though about contacting Machintosh? Make it a
Carbon question, not an emacs question.
Cheers,
-- Nikodemus
·········@yahoo.co.uk (BK) writes:
> Nikodemus Siivola <········@kekkonen.cs.hut.fi> wrote ...
>
> > Just a thought: are the control and meta available to the carbon emacs at
> > all?
>
> yes, some combinations work - some don't
>
> I am now trying yet another build and see how that one goes ;-)
>
> > Whenever there has been ctrl/meta troubles on my boxes it has always
> > turned out to be a case of the window manager hogging them...
>
> interesting. could it "hog" just some combinations but not others?
Check the Full Keyboard Access tab in the Keyboard System Preferences.
If it's on and you're using "Letter Keys" you may be stealing five
C-<key> combos. I was pretty mystified for a while why C-g didn't work
while lots of other key chords did.
-Peter
P.S. My Mac isn't my main computer so I don't have that much
experience with actually making Emacs, etc. on it work. But I did run
into this.
--
Peter Seibel ·····@javamonkey.com
The intellectual level needed for system design is in general
grossly underestimated. I am convinced more than ever that this
type of work is very difficult and that every effort to do it with
other than the best people is doomed to either failure or moderate
success at enormous expense. --Edsger Dijkstra
Nikodemus Siivola <········@kekkonen.cs.hut.fi> writes:
> David Steuber <·············@verizon.net> wrote:
>
> > I have no idea why you are having trouble with the meta and ctrl keys
> > in carbon emacs. I haven't figured out the deal with GNU emacs yet
> > ;-). I imagine the Mac just twiddles different bits.
>
> Just a thought: are the control and meta available to the carbon emacs at
> all?
>
> Whenever there has been ctrl/meta troubles on my boxes it has always
> turned out to be a case of the window manager hogging them...
I would certainly expect the Mac keyboard to generate different
scancodes for different keys. Then there is all the software layers.
But if Carbon Emacs works at all, then there must be a mapping setup.
It should have been handled by the porters. I can't see them not
using the program at all. Eat your own dog food, ya know?
However, I do get slightly different behavior if I am on my Debian box
rather than sshing in from my Mac. But that is with GNU.
As for that silly Free vs Commercial argument, GNU Emacs is free. It
seems pretty popular and wide spread if "you get what you pay for."
--
(describe 'describe)
On 7 May 2003 13:37:14 -0700, ·········@yahoo.co.uk (BK) wrote:
> How is Lisp going to win more friends if you have to fight with the
> editor for a week?! That will scare away anyone but the most
> determined. Perhaps the old-hands in the Lisp community like it that
> way, I don't know. Perhaps the Lisp community is an elitist club and
> they want nobody but the most determined to join them, I don't know.
Concerning ILISP, I can assure you that the explanation is very simple and
mundane: no elitism, just limited resources.
Paolo
--
Paolo Amoroso <·······@mclink.it>
On 7 May 2003 13:37:14 -0700, ·········@yahoo.co.uk (BK) wrote:
> How is Lisp going to win more friends if you have to fight with the
> editor for a week?! That will scare away anyone but the most
> determined. Perhaps the old-hands in the Lisp community like it that
> way, I don't know. Perhaps the Lisp community is an elitist club and
> they want nobody but the most determined to join them, I don't know.
Several thoughts come to mind reading the above. I belive the problem
here is not with Lisp but with free software in general.
You are using a new configuration on a new system. That is asking for
trouble. On the plus side, you are likely to be in pretty good
company, intellect-wise, but on the minus side, it is very likely that
you will have to figure out a lot of stuff yourself.
One of the things that always perplexes me is why people expect free
software to be bug-free, easy to install, and trouble free to use. In
general, free software is useless, or to put it another way, there are
far more useless or only marginally useful pieces of free software out
there than useful ones. Many pieces of free software proudly claim to
be guaranteed buggy, and say that if you happen to be able to use it
for anything you should consider yourself extremely lucky. Others,
wanting to maintain more of an appearance of professionalism, simply
"disclaim all warrantee", but this amounts to much the same thing,
from the point of view of recourse on the user's end if it doesn't
work.
Of course the commercial side of things is not much better, since the
EULAs of most commercial software say something similar --- and they
make you pay for it. But at any rate if you pay for something you
should at least get ranting-rights. (Microsoft would even like to
take this from you with their new "you swear, we sue" EULA, but I have
to wonder whether that would stand up in court. IANAL)
What I'm trying to say here in a kind of sarcastic way is that part of
the beauty of free software is that you don't have to pay for support
if you don't need it. But the flip side is that if you need support,
you may have to pay for it to get it, at least at the level you would
like. If you can't or aren't willing to pay, you should expect to
ramp yourself up into being able to take care of yourself in the
support area, with possible help from online resources. If you can't
do that, it really is your problem.
At best, you should at least divest yourself of the sense of
entitlement that I allude to in my third paragraph above. Yes, many
of us have gotten this stuff to work. What is not apparent is the
hours of time that went into it, including the expertise we developed
from installing and debugging other software over the years. (See my
signature below.) If you can benefit from that experience without
having to go through the pain of gaining it, so much the better. But
you should not expect that by right.
Yes, it may take a week of fighting with the editor to get it to work
the way you want. You can see that as purely negative, or you can see
it as a positive thing. First you may, if you can communicate your
experience, save someone else that week. Second, you may be able to
contribute intelligent bug reports and positive suggestions for
improvement that will make the software better. And third, you have
learned something that will make it easier next time. If you aren't
willing to go through this from time to time, you are really a
free-rider, and, to be blunt, you should be grateful for what you get.
My own approach is that I'm willing to ask stupid questions, but I
will never criticize or complain about software I got for free. If I
don't like it, and can't make it work the way I want, I just won't use
it. If I submit a bug report, I always assume that I may be the one
at fault --- because a significant part of the time I *am* at fault.
I still remember many years ago reporting a bug in a public forum
where I'd run the wrong version of a program, and was claiming a bug
still existed that was fixed, and I convinced the author that the bug
still existed. I later had to explain my mistake, and the author
wasn't that happy about it. But at least I'd only looked stupid, not
like a complete clueless idiot, because I was polite about it all the
way through.
--
Fred Gilham ······@csl.sri.com
"I'm an expert at installing free software. I've installed software
packages that had 412 steps, the first two of which were 'Remove small
children and animals from the premises' and 'Don protective gloves and
mask'. If you made a mistake you had to go back to the very
beginning, including getting the kids and pets back in the house."
Hi Fred Gilham,
> On 7 May 2003 13:37:14 -0700, ·········@yahoo.co.uk (BK) wrote:
>> How is Lisp going to win more friends if you have to fight with the
>> editor for a week?! That will scare away anyone but the most
>> determined. Perhaps the old-hands in the Lisp community like it that
>> way, I don't know. Perhaps the Lisp community is an elitist club and
>> they want nobody but the most determined to join them, I don't know.
>
> Several thoughts come to mind reading the above. I believe the problem
> here is not with Lisp but with free software in general.
>
> You are using a new configuration on a new system. That is asking for
> trouble. On the plus side, you are likely to be in pretty good company,
> intellect-wise, but on the minus side, it is very likely that you will
> have to figure out a lot of stuff yourself.
>
> One of the things that always perplexes me is why people expect free
> software to be bug-free, easy to install, and trouble free to use. In
> general, free software is useless, or to put it another way, there are
> far more useless or only marginally useful pieces of free software out
> there than useful ones.
I genuinely expect free software to be relatively bug free (at least as
bug free as proprietary software), easy to install (easier than
proprietary software and many times easier to update) and relatively
trouble free to use (at least as easy to use as proprietary software, in
the spirit of "Contrary to popular belief, UNIX is a user-friendly
Operating System. It's just choosey about who its friends are.")
And that's the case with ILISP. Typing "apt-get install ilisp" wasn't
hard. And working with ILISP/Debian developers to resolve a bug where Lisp
in used in the ANSI-specified case-preserving invert mode wasn't hard
either.
The Debian development model and its distribution tracks in general
address the problems "with free software in general". As do other
distributions in similar and different ways (e.g. commercial distributors
that provide the option to purchase high quality software support).
I agree that one is going to face significant problems when "using a new
configuration on a new system. That is asking for trouble." This is the
problem that free software distributions seek to address. It's not a
problem with free software in general.
Regards,
Adam
"Adam Warner" <······@consulting.net.nz> wrote:
> > One of the things that always perplexes me is why people expect
> > free software to be bug-free, easy to install, and trouble free to
> > use. In general, free software is useless, or to put it another
> > way, there are far more useless or only marginally useful pieces
> > of free software out there than useful ones.
>
> I genuinely expect free software to be relatively bug free (at least
> as bug free as proprietary software), easy to install (easier than
> proprietary software and many times easier to update) and relatively
> trouble free to use (at least as easy to use as proprietary
> software, in the spirit of "Contrary to popular belief, UNIX is a
> user-friendly Operating System. It's just choosey about who its
> friends are.")
>
> And that's the case with ILISP. Typing "apt-get install ilisp" wasn't
> hard. And working with ILISP/Debian developers to resolve a bug where Lisp
> in used in the ANSI-specified case-preserving invert mode wasn't hard
> either.
Sometimes it works, sometimes it doesn't.
boomerang:~ > apt-get install ilisp
apt-get: Command not found.
boomerang:~ >
See, I told you!!!!
:-)
I've decided to start the Trotskyist Software Movement. It will
create something called the TPL (Trotskyist Program License). It will
go as follows:
Preamble
This license is for programmers in the forefront of the
revolutionary struggle. It attempts to protect no rights.
Rights are a bourgeois concept based on a religiously inspired
moral view that has opiated the masses, especially the
programming masses, for far too long. The intent of this license
is that programmers should program! Long live the struggle, and
may it soon be extended to all operating systems!
Once one has finished a piece of software, a deleterious process
takes place. Does the author immediately proceed to write
another piece of software? No! He sits down and thinks, "How
can I protect my rights? How can I profit? Or how, at least,
can I gain credit for my work?" He may consider several
different software licenses. "Perhaps I should put it under the
GPL. Maybe I can sell it!"
And all this time, he could have been coding! What waste, what
inefficiency! How can someone be a programmer unless he
programs! How can someone be a hacker if he does not hack!
And hack indeed is the word. For each program written is an
axe-cut at the root of the oppressive capitalist tree, the dying
husk of which dominates the otherwise fertile ground, drawing
nutrient that could otherwise go to a vast crop of production.
Needs could be met, the proletariat fed, the workers paradise
become a reality --- but only if each person sees himself
personally at the front of the struggle!
In programming, as in nature itself, there is no final product.
Today we have version 1.0, tomorrow, 1.1; perhaps, in a glorious
expression of revolutionary joy, we go to 2.0 and cause all users
to start from the beginning! There is always another version,
always another release! And this illustrates, nay IS, the
revolutionary struggle. Only at the end of time will it cease,
as Moore's law finally becomes null with the heat-death of the
universe, and the heroic hacker can lean back and admire a job
well done.
Until then, though, the struggle continues. The capitalist
thieves may steal our work. The small minded may scurry around
trying to protect it. We laugh! Our work can protect itself!
And finally, we rejoice in our revolutionary motto, which is,
"There's more where that came from!"
I suspect the above will quickly make all other "free" software
licenses obsolete.
--
Fred Gilham ······@csl.sri.com
Is, then, the loving cup so often filled
that we may toss a draught aside?....
-Jeff Iverson
Hi Fred Gilham,
> "Adam Warner" <······@consulting.net.nz> wrote:
>> > One of the things that always perplexes me is why people expect
>> > free software to be bug-free, easy to install, and trouble free to
>> > use. In general, free software is useless, or to put it another
>> > way, there are far more useless or only marginally useful pieces
>> > of free software out there than useful ones.
>>
>> I genuinely expect free software to be relatively bug free (at least
>> as bug free as proprietary software), easy to install (easier than
>> proprietary software and many times easier to update) and relatively
>> trouble free to use (at least as easy to use as proprietary
>> software, in the spirit of "Contrary to popular belief, UNIX is a
>> user-friendly Operating System. It's just choosey about who its
>> friends are.")
>>
>> And that's the case with ILISP. Typing "apt-get install ilisp" wasn't
>> hard. And working with ILISP/Debian developers to resolve a bug where Lisp
>> in used in the ANSI-specified case-preserving invert mode wasn't hard
>> either.
>
> Sometimes it works, sometimes it doesn't.
>
> boomerang:~ > apt-get install ilisp
> apt-get: Command not found.
> boomerang:~ >
>
> See, I told you!!!!
>
> :-)
>
>
> I've decided to start the Trotskyist Software Movement. It will
> create something called the TPL (Trotskyist Program License). It will
> go as follows:
(snipped very funny preamble)
I'm just wondering whether I'm in on the joke or the subject of the joke
(if it's the latter I'd ask you to reevaluate my response). My expectations
come from my experience in using Debian GNU/Linux for a number of years.
It's also a Lisp-friendly platform where it is easy to install most Lisp
software (look for example at the packages Kevin Rosenberg maintains for
Debian at <http://b9.com/debian.html>. The first item on the list even
shows how free Debian packages can assist in the download and installation
of proprietary software like ACL. One day Franz may even decide it is in
their interests to package ACL for Debian and provide an update service
for their customers using the official package system. They could do this
because the update system is fully under user control [many update systems
are controlled by vendors who act as a gatekeeper to what software may be
distributed via the official update system]).
Regards,
Adam
> I'm just wondering whether I'm in on the joke or the subject of the joke
> (if it's the latter I'd ask you to reevaluate my response).
I hope you're in on it. Feel free to send me software written under
the TPL! :-)
Actually I sort of conflated two threads in my mind and wound up with
what I posted....
What has me worried is that someone will come up with something like
the RTPL (Revolutionary Trotskyist Program License) or the PTPL
(People's Trotskyist Program license), the first of which takes issue
with my comment about the heat death of the universe (revolutionary
struggle should seek to prevent it) and the second of which sees
hackers as elitist and believes that all programs should be written by
the masses flailing away at millions of keyboards. You know how we
revolutionaries are!
On a serious note, I have believed that software should be open source
from the days I first turned on my H8 micro and watched it scroll
"your h8 is up and running" on the front panel lights. While in grad
school I read a convincing argument that programmers function better
in a source-rich environment. Even Bill Gates himself became a good
programmer by rifling through dumpsters to find program listings and
reading them, or so the legend goes.
I personally don't mind a bit of struggle :-) with installing free
software (see my signature below, which came up completely at
random!). But I know that the movement as a whole gains credibility
when it can produce high quality software that competes with
commercially written software. What I take issue with is the sense of
entitlement that sometimes arises in people who use it.
--
Fred Gilham ······@csl.sri.com
"I'm an expert at installing free software. I've installed software
packages that had 412 steps, the first two of which were 'Remove small
children and animals from the premises' and 'Don protective gloves and
mask'. If you made a mistake you had to go back to the very
beginning, including getting the kids and pets back in the house."
Paolo Amoroso <·······@mclink.it> wrote ...
> Concerning ILISP, I can assure you that the explanation is very simple and
> mundane: no elitism, just limited resources.
Fair enough. And once I am done, you will have an easy to understand
section (or README) for Mac users on how to get it working when using
Aqua ;-)
Trouble right now is on the Emacs side of things, not ILISP. Perhaps,
Emacs on Aqua hasn't reached the same maturity as XEmacs yet.
rgds
bk
·········@yahoo.co.uk (BK) writes:
> How is Lisp going to win more friends if you have to fight with the
> editor for a week?! That will scare away anyone but the most
> determined. Perhaps the old-hands in the Lisp community like it that
> way, I don't know. Perhaps the Lisp community is an elitist club and
> they want nobody but the most determined to join them, I don't know.
Just don't fight the editor. If ilisp makes you cringe, don't use
it. Instead:
Emacs comes with a nice lisp mode, fairly usable, that fires up
automagically whenever you load a .lisp file.
When you do M-x run-lisp, emacs tries to run a lisp on your system,
and if it finds it, it runs it in a buffer called *inferior-lisp*, in
which you can type in commands, or to which you can send whole
expressions from your .lisp file via C-c C-c.
Your lisp compiler/interpreter should be called "lisp", and be
reachable through some path in PATH.
When editing a lisp file (this trick actually works everywhere else),
you can type C-h b and get a window with half a ton of keybindings to
go "wow!" trying them out for about a week, instead of sitting there
contemplating on which directory might be the Optimum Choice for
installing Ilisp.
You could of course install XEmacs, which AFAICT comes with ilisp
preinstalled.
Maybe it would be better if people here stopped to recomend ilisp to
every newbie. That thing just doesn't seem to be plug and play.
Mario.
> Just don't fight the editor. If ilisp makes you cringe, don't use
> it. Instead:
>
> Emacs comes with a nice lisp mode, fairly usable, that fires up
> automagically whenever you load a .lisp file.
[snip]
> Maybe it would be better if people here stopped to recomend ilisp to
> every newbie. That thing just doesn't seem to be plug and play.
While I've used ILISP in the past, and will probably use it in the
future, I haven't bothered to use it for quite some time now, and have
been perfectly happy with the standard lisp support (in GNU Emacs). So
I'm definitely sympathetic to this idea.
Ted
--
Edward O'Connor
·······@soe.ucsd.edu
Edward O'Connor <·······@soe.ucsd.edu> wrote ...
> > Just don't fight the editor. If ilisp makes you cringe, don't use
> > it. Instead:
> >
> > Emacs comes with a nice lisp mode, fairly usable, that fires up
> > automagically whenever you load a .lisp file.
> [snip]
> > Maybe it would be better if people here stopped to recomend ilisp to
> > every newbie. That thing just doesn't seem to be plug and play.
>
> While I've used ILISP in the past, and will probably use it in the
> future, I haven't bothered to use it for quite some time now, and have
> been perfectly happy with the standard lisp support (in GNU Emacs). So
> I'm definitely sympathetic to this idea.
Thanks for your concerns, but please allow me to defend ILISP here.
What I am having a very hard time with is clearly on the side of
Emacs, carbonised Emacs to be precise, and not ILISP.
I think ILISP is a nice extension to Emacs if you want to do Lisp and
once you have got over the initial hurdle of installation it doesn't
really seem to be causing any trouble.
The difficulties I had getting ILISP to work are due to specifics of
the Aqua environment.
If you want to work under Aqua and use an Aqua version of Emacs, you
have to figure out how to apply the instructions in the ILISP
documentation to the Aqua environment. I have documented that already
and posted a detailed summary follow up. The difficulties have to do
with the Aqua environment, not ILISP's fault, but the ILISP
documentation doesn't deal with that.
Thus, it would be a good idea to have a section in the ILISP
documentation that describes how to go about setting up Emacs and
ILISP for Aqua. As other posters have pointed out, this is all free
software and there is no entitlement to such documentation. One may
therefore just take the view "Tough luck - you work it out for
yourself" and there is nothing wrong with that in principle.
However, I am asking myself this. As I am getting this for free, if I
am going through all that trouble, spending all that time, why do I
not give something back and write up a section for the documentation
that describes this special case step by step so that others who come
after me can be spared having to go through the same trouble all over
again?
I would also emphasise that Mac users are used to having it a bit
easier and not everybody is able or willing to go through all that.
You may say that if they have this attitude they shouldn't be able to
use the software in the first place. But I say this would be a loss
for the community. I believe that the expectation of Mac users to get
started right away is a positive thing. I don't see this as a sign of
laziness. I see it as a sign of efficiency. Mac users are willing to
work as hard as anyone else, but they don't see the point of spending
their time on things that don't really have anything to so with what
they actually set out to do, in this case learning or working with
Lisp.
I believe that this spirit is an enrichment for the computer industry.
The more users are demanding that they don't have to be system admins
to use even free software, the more user friendly environments will
become and the more time people can spend on their own work or study.
It will be a benefit for everyone. I am not a good programmer and I
probably never will be, for me doing Lisp is more fun than anything
else. But if my writing down step-by-step what to do in order to get
started with Emacs and ILISP under Aqua helps to get only *one* young
kid with talent to play with Lisp and stick with it instead of tossing
it and turn attention to C++, then it will have been worth the
trouble. There is always the chance there is another Guy Steele
amongst those kids. We wouldn't want to lose them, would we?!
So, as I am trying to complete/refine the section how to make this all
work, I try out things and I see that none of the Emacs builds do what
the Emacs documentation says they should do. I wouldn't mind if those
things are advanced features, but its the most elementary things like
quitting and pasting and most troublesome, the keyboard shortcut to
invoke the HyperSpec (via ILISP) is deaf.
If I am to write up a section how to make it work and part of that is
the HyperSpec, I cannot just ignore this, can I? So, I have to fight
with Emacs and see how it can be put into a usable state. I have
started this and I am going to finish it, somehow.
rgds
bk
·········@yahoo.co.uk (BK) writes:
> If I am to write up a section how to make it work and part of that is
> the HyperSpec, I cannot just ignore this, can I? So, I have to fight
> with Emacs and see how it can be put into a usable state. I have
> started this and I am going to finish it, somehow.
if you have to fight w/ emacs you will find you will have to fight w/
those people who are comfortable w/ emacs. generally, fighting is a
dissipative luxury not afforded by poor hackers w/ limited lifespan.
thi
Thien-Thi Nguyen <···@glug.org> wrote ...
> if you have to fight w/ emacs you will find you will have to fight w/
> those people who are comfortable w/ emacs. generally, fighting is a
> dissipative luxury not afforded by poor hackers w/ limited lifespan.
that's what I figured, too :-)
thanks
rgds
bk
·········@yahoo.co.uk (BK) writes:
> How is Lisp going to win more friends if you have to fight with the
> editor for a week?! That will scare away anyone but the most
> determined. Perhaps the old-hands in the Lisp community like it that
> way, I don't know. Perhaps the Lisp community is an elitist club and
> they want nobody but the most determined to join them, I don't know.
I once made the error to recommend Linux and Emacs to members of a
programming course using C++. Let it suffice to say that I won't do that
again. Now, I am giving an introductory course "AI programming in CL"
(where I want to do part of Norvig's PAIP). For the exercises, I am using
the trial edition of Lispworks. The course has just started, but
apparently using a commercial trial edition is by far the best way to get
started with CL.
Nicolas.
Nicolas Neuss <·············@iwr.uni-heidelberg.de> writes:
> I once made the error to recommend Linux and Emacs to members of a
> programming course using C++.
Heh. I got used to C++ in a series of very nice and steadily
improving Windows IDEs. It would have been bad for me if I had to
first go through the overhead of learning the GNU tools as well. As
it is, I still haven't mastered them on Linux. At least there is Perl
;-)
Mind you, there are plenty of Unix shops that use GNU even though
there are some very impressive proprietary tools available. Nothing
beats having a good debugger. I would say that not having to write
make files is a nice thing though. YUCK!
--
(describe 'describe)