········@gmail.com wrote:
> It has been repeatedly and voluminously proposed that a cadre or cabal
> be formed to create a neolisp sublanguage -- a sanctioned set of
> add-ons (NOT replacements) for Common Lisp. (I prefer the term CADRe
> for obvious reasons.) I tend to agree with those who think that this
> needs to be done in public by a semi-formal elected committee. But I'm
> not much for talk. Therefore, I propose that we simply form the
> committee and start putting together the addons.
Just for the enlightenment of the clueless, what happened to this
effort? http://clrfi.alu.org/
I'd agree we need something like this (I suppose I'm part of the
voluminous proposals you refer to) but it needs to be well structured,
well organized, and have the active support of enough of the Lisp
community to have moral authority. Probably the best way is to create
good proposals. Also, perhaps the commercial Lisp vendors would have
some interest in such a process, as well as the resources to do high
quality work?
"C Y" <···········@yahoo.com> writes:
> Just for the enlightenment of the clueless, what happened to this
> effort? http://clrfi.alu.org/
Some people took the CLRFI idea and tried to define a process which
would give them work, which they didn't do, so the process stopped.
In my opinion, in this time of web and wiki, the CLRFI could just be
posted and commented in cliki.net, without a need for a central
"authority", at least until some momentum is gained and resources
agregated.
For example, useful libraries like CFFI could be posted as a CLRFI.
--
__Pascal Bourguignon__ http://www.informatimago.com/
PUBLIC NOTICE AS REQUIRED BY LAW: Any use of this product, in any
manner whatsoever, will increase the amount of disorder in the
universe. Although no liability is implied herein, the consumer is
warned that this process will ultimately lead to the heat death of
the universe.
Pascal Bourguignon <···@informatimago.com> writes:
> For example, useful libraries like CFFI could be posted as a CLRFI.
They could, but I think that this would be a bad idea: libraries like
CFFI are unstable in the sense that their interface is changing as new
requirements are being discovered. What might make more sense to
submit as a CLRFI is some interface to the low-level operations that
CFFI uses to implement itself: operators dealing with things like C
pointers, pinning vectors against being moved by garbage collectors,
and similar. Bonus points if these operators are required by more
than one 'userspace' library, so that one can judge the semantics
needed against more than one use case.
Christophe
Christophe Rhodes wrote:
> They could, but I think that this would be a bad idea: libraries like
> CFFI are unstable in the sense that their interface is changing as new
> requirements are being discovered. What might make more sense to
> submit as a CLRFI is some interface to the low-level operations that
> CFFI uses to implement itself: operators dealing with things like C
> pointers, pinning vectors against being moved by garbage collectors,
> and similar. Bonus points if these operators are required by more
> than one 'userspace' library, so that one can judge the semantics
> needed against more than one use case.
I think the problem with that idea is that implementations will already
have these sorts of things, and they will all be different, and the
implementors will have a significant investment in how they do things
now (lots of working code) and won't want to change them, quite
reasonably. Specifying some thin standard wrapper around the various
extant implementations might be a good approach, but only if they are
similar enough that one or more is not unduly penalised. This matters
a lot for FFIs in particular of course - no one wants some standard
wrapper that causes enormous overhead when calling C code, on *some*
implementations.
So I think a higher-level standard interface would be a better idea.
But if CFFI is unstable then that's probably a good indication that the
time is not yet ripe for one.
Of course, there are things that could be done that aren't this hard.
For instance (just from looking at the CFFI-SYS manual and cringing):
how hard would some package-related standards be? It is about 10 years
past time that people started giving packages sensible names and it's
easy to just borrow what Java has done there (ORG.TFEB.UTILS for
instance, not UTILS). Then we want things like short names for
packages, and there are some fairly obvious things to do with that.
And all of this has the nice property that, though implementation
changes will be required, they will probably be very small ones, and
also not on any kind of critical performance path (really, you just
need a few hooks on which to hang the new DEFPACKAGE etc). Finally,
getting this done lets you construct variant CLs by package
manipulations of various kinds (for instance: one where the symbol you
get buy reading CL:DEFPACKAGE is not the common lisp DEFPACKAGE. So
something like this is perhaps a good jumping-off place.
(Of course, I happen to have done work implementing things that do all
this, which is why I'm mentioning it - there are probably other areas
which would pay off at least as well. In any case what I *don't* have
is a specification, I just have code. Still less do I have a
specification which might be acceptable to implementors.)
I'm not disagreeing with you by the way - at least not intentionally.
On the subject of BDFLs which came up in another branch of this thread.
All the mentioned BDFLs are for *implementations* not for standards.
Standards are very different things, and I'm not aware of any driven by
BDFLs. In many ways standards are *harder* than implementations,
because for them to be useful they need to be widely accepted, and that
requires social skills. Which is why the coup d'etat approach is not
likely to work for them.
Finally, the money issue. Writing code is something that people will do
for free (fools that they are), because it has an immediate payoff -
you have a problem, so you write something that makes the problem go
away. And then you can write a little bit more code and some more
problems go away - it's just drugs without the being-illegal bit
really. Writing standards, and doing all the political stuff involved
with them is a different issue: there is no immediate hit from solving
a problem, and you have to deal with a whole lot of social & political
stuff that most programmers find very painful, are really bad at and
avoid whenever they can. And writing good standards is also just hard
work, even without the human interaction aspects, requiring very good
command of language and so on. My guess is that very few standards are
written by people for free, and those that are will generally be thin
skins on an imlementation (like the python standard or something). No
one is going to work on POSIX for free for instance! And though the
original poster has mentioned money, I don't think he realises how
little it is, given that people with the skills to do standardisation
work also have the skills to earn a lot of money (all those social
skills turn out to matter).
--tim
Tim Bradshaw wrote:
> ... though the original poster has mentioned money, I don't think
> he realises how little it is ...
Of course he relalizes how little it is. He didn't just offer the cash,
he challenged others to step up like he did. You don't have to give
$10,000 -- how about $1000? How about $500? How about $100?
········@gmail.com wrote:
> Of course he relalizes how little it is. He didn't just offer the cash,
> he challenged others to step up like he did. You don't have to give
> $10,000 -- how about $1000? How about $500? How about $100?
If the goal is newbie adoption, it sounds like there's a much easier
solution, based on work that's already been done for us by hard
workers. The canonical newbie distro can be LispBox. The nice website
can be built around it, with recommended, tested libs. Resources can be
applied to port things to Wintel and MacOS X.
If someone else thinks this is a bad approach, they can do something
else. A democratic process where people manage themselves.
And really, it seems more important to hear newbies talk, rather than
experienced Lisp users debate. Seems the newbies' problems are much
more humble than we think. Like being able to reliably install CLSQL
under Lispbox. (Which I kinda did yesterday, and had a big hiccup,
which would've turned off the casual newbie. I say "kinda" because I
didn't use Lispbox but did use asdf-extensions.lisp which I extracted
from Lispbox. That interacts badly with makefiles, requiring a
workaround; an issue I expect Lispbox users face.)
Tayssir
> If the goal is newbie adoption, it sounds like there's a much easier
> solution, based on work that's already been done for us by hard
> workers.
The goal isn't newbie adoption, but if it were, you'd be right, Lispbox
is good. Regardless, what you are in fact exactly right about is that
building and maintaining Lispbox is hard work done by hard workers, and
the guy/folks who does/do it need to pay his/their rent and
occasionally eat. As has been pointed out, $10,000 isn't much, but it's
something. I don't mean this in a nasty way, but I'll bet you and the
rest of the community aren't going to offer the Lispbox guy/folks a
dime for all his/their hard work, nor will you (that is, the community)
offer him/them, nor anyone else anything but a warm handshake for the
work involved in doing what it takes to keep Lisp alive. The open
source/freeware mentality has broken something deep in the world, and
it's killing Lisp, and probably lots of other good things. Companies
that build fantastic Lisp platforms can barely stay in business and
need to scratch for every dime to pay their hard-working engineers, and
the community is full of people who think that just because they write
and give away yet another crappy, partly working implementation of
variable persistence that they knocked off in their spare time, they
deserve to get excellent working software by the hard work of brilliant
people for free. Ugh.
(Sorry, TJG, this flame isn't directed at you -- just prompted by your
note.)
········@gmail.com wrote:
> I don't mean this in a nasty way, but I'll bet you and the
> rest of the community aren't going to offer the Lispbox guy/folks a
> dime for all his/their hard work, nor will you (that is, the community)
> offer him/them, nor anyone else anything but a warm handshake for the
> work involved in doing what it takes to keep Lisp alive.
My reply may be besides your point but I'd like to offer, as a
data-point, that part of the community collectively gave $1123.09 to
common-lisp.net. If you look at the donors list you'll recognize some
of them as regular posters to this group. While it may not be much
money, the annual budget is only $900.
(Thanks everyone!)
Erik.
········@gmail.com wrote:
> (Sorry, TJG, this flame isn't directed at you -- just prompted by your
> note.)
No problem. I personally feel the recent discussions here are almost
too objectionable and irrational to participate in, partly because I
don't think people are interested in listening to each other
(preferring debate rather than building something). While I may not
agree with your points, it's not a reason for aiming abuse at you.
> > If the goal is newbie adoption, it sounds like there's a much easier
> > solution, based on work that's already been done for us by hard
> > workers.
>
> The goal isn't newbie adoption, but if it were, you'd be right, Lispbox
> is good.
Hmm, what's your goal? I (and perhaps others) haven't understood you,
in that case.
> Regardless, what you are in fact exactly right about is that
> building and maintaining Lispbox is hard work done by hard workers, and
> the guy/folks who does/do it need to pay his/their rent and
> occasionally eat. As has been pointed out, $10,000 isn't much, but it's
> something. I don't mean this in a nasty way, but I'll bet you and the
> rest of the community aren't going to offer the Lispbox guy/folks a
> dime for all his/their hard work, nor will you (that is, the community)
> offer him/them, nor anyone else anything but a warm handshake for the
> work involved in doing what it takes to keep Lisp alive.
Yes, there isn't even much funding for the people working hard on vital
issues like environmental destruction and need for independent media,
is there? Still, we prioritize. Where money is effective, maybe we give
money. Where our skills as Lisp programmers are effective, maybe we
contribute in that manner.
Or not.
> The open
> source/freeware mentality has broken something deep in the world, and
> it's killing Lisp, and probably lots of other good things. Companies
> that build fantastic Lisp platforms can barely stay in business and
> need to scratch for every dime to pay their hard-working engineers, and
Such is our economic system... it does not support infinitely
reproduceable goods well.
> the community is full of people who think that just because they write
> and give away yet another crappy, partly working implementation of
> variable persistence that they knocked off in their spare time, they
> deserve to get excellent working software by the hard work of brilliant
> people for free. Ugh.
Which persistence libs are you critiquing? A serious critique sounds
interesting.
Tayssir
> > > If the goal is newbie adoption, it sounds like there's a much easier
> > > solution, based on work that's already been done for us by hard
> > > workers.
> > The goal isn't newbie adoption, but if it were, you'd be right, Lispbox
> > is good.
> Hmm, what's your goal? I (and perhaps others) haven't understood you,
> in that case.
It's having a systematic process through which the langauge -- by which
I mean the core language and the "standard library set" which amounts
to a distributed part of the language -- can be extended to encompass
modern needs and discoveries in order that professional work can be
done in it, and that we (professionals) can expect (and may have to
expect to support!) the continued growth of the language as new needs
and discoveries are discovered. The goal -- my goal, anyhow -- is to
retain professionals, and lead professionals to choose lisp for app
development over, say, Python, PERL, or Java or even C/C++ (although I
think that the latter serve a useful role in the world at the moment in
deep near-hardware development). I'm not just pissing this stuff: I
program in Lisp every single day of the week, and have to nearly 30
years. I do absolutely everything in Lisp (except, as above,
near-hardware stuff, and in many cases I do that with Lisp-based
macros!) I've written god knows how much code, and published over 50
peer-reviewed professional papers most of who's contents is ENTIRELY
based on Lisp -- InterLisp, *Lisp, MacLisp, CMUCL (the origianl -- I
was at CMU!). But, you know, every time I need to dig around to find
something that is essentially standard in Python or Java, and end up
just writing it myself and hating it, I have to resist the urge to jump
ship myself. And it's not merely that the language doesn't have X or Y,
it's that there is no process by which X or Y could EVER become part of
the language (or standard libraries), and so I watch Python, Perl, C#,
and Java racing by while I'm yet-again writing my own dumb partial
implementation of something that should be standard in ANY MODERN
LANGUAGE. The language is like Latin. It's perfect for the Roman world.
And like Latin it's perfectly dead. So, to answer your quesiton: My
goals is not to attract newbies, it's to keep myself from breaking my
own heart by leaving the love of my life.
········@gmail.com wrote:
>>>>If the goal is newbie adoption, it sounds like there's a much easier
>>>>solution, based on work that's already been done for us by hard
>>>>workers.
>>>
>>>The goal isn't newbie adoption, but if it were, you'd be right, Lispbox
>>>is good.
>>
>>Hmm, what's your goal? I (and perhaps others) haven't understood you,
>>in that case.
>
>
> It's having a systematic process through which the langauge -- by which
> I mean the core language and the "standard library set" which amounts
> to a distributed part of the language -- can be extended to encompass
> modern needs and discoveries in order that professional work can be
> done in it, and that we (professionals) can expect (and may have to
> expect to support!) the continued growth of the language as new needs
> and discoveries are discovered.
This is a great goal and I wish you luck, notwithstanding the constant
naysayers on this group who come up with the same tired non-logic and
whine for a specific example, then desparately attack it rather than
address the lack of process for change anywhere. I've lobbied for change
here in the past. Kent Pitman wanted to address it with his
Substandards, various luminaries tried again with CLRFI, and a few
former lurkers often come out in favour of progress every six months
when this issue comes up.
> The language is like Latin. It's perfect for the Roman world.
> And like Latin it's perfectly dead. So, to answer your quesiton: My
> goals is not to attract newbies, it's to keep myself from breaking my
> own heart by leaving the love of my life.
This is, unfortunately, not true. Latin is alive and well and has been
updated more recently than Common Lisp. Which just goes to show that if
you reduce change to absolute zero and hold it there long enough, even
the Roman Catholic Church will eventually speed by you.
http://www.economist.com/diversions/displayStory.cfm?story_id=2281926
On 20 Apr 2006 09:41:48 -0700, ········@gmail.com wrote:
>
> It's having a systematic process through which the langauge -- by which
> I mean the core language and the "standard library set" which amounts
> to a distributed part of the language -- can be extended to encompass
> modern needs and discoveries in order that professional work can be
> done in it, and that we (professionals) can expect (and may have to
> expect to support!) the continued growth of the language as new needs
> and discoveries are discovered. [snip snip]
In my opinion, Common Lisp is caught in a catch-22. On the one hand,
the language and the standard library are not a commodity. Thus in
the marketplace for implementations and tools, customer lock-in is
still a viable business plan -- given all the parameters that go
into a decision of which implementation(s) to use, today the path
of least resistance given many permutations of those parameters is
to just go with a single vendor, and then it becomes very, very
hard to migrate to a different implementation. This situation is
facilitated in part by the spec that we have today (which we must
recognize is a product of not just a technical process, but also a
political process).
The vendors today have enough marketshare and remain healthy enough
as businesses that the bar for commoditization to take hold is too
high. The FOSS implementations are strong in their own right, but
not yet compelling enough to displace the commercial implementations.
And the commercial vendors see a relatively low potential ROI
in encouraging further standardization. So for all of these
reasons, the status quo remains. But I do think there is a
change occurring.
I do not in any way blame the commercial vendors for finding
strategies to earn revenue. I would be hard-pressed to come up
with better strategies, and I deeply respect the ability of these
companies to survive conditions that would have put lesser companies
out of business long ago. In other words, I don't have an
anti-commercial mindset. I'm just explaining my view as to
why a sanctioned standards process is a non-starter right
now.
So, my thinking is that Common Lisp can only evolve at this
point through ad hoc standards, with CFFI being the easiest
and most obvious example of an emerging ad hoc standard.
It's getting harder and harder to justify staying with
vendor-specific FFI. To my way of thinking, emerging ad hoc
standards are the most likely disruptive development that
could change the status quo.
Obviously I can't say where the tipping point is, but I do
believe that there is a tipping point where commoditization
can occur. That's where I see a sanctioned process coming
into play, and in my opinion a process like that would become
essential.
--
Jack Unrue
········@gmail.com writes:
> was at CMU!). But, you know, every time I need to dig around to find
> something that is essentially standard in Python or Java, and end up
> just writing it myself and hating it, I have to resist the urge to jump
> ship myself. And it's not merely that the language doesn't have X or Y,
> it's that there is no process by which X or Y could EVER become part of
> the language (or standard libraries), and so I watch Python, Perl, C#,
> and Java racing by while I'm yet-again writing my own dumb partial
> implementation of something that should be standard in ANY MODERN
> LANGUAGE. [...]
If it is any good, why don't you post your code for X and Y on
your web site or create projects on common-lisp.net? It very well
could become part of the standard libraries in the same way that, I would
assert, asdf, cffi, clx and many others are now.
Tim
········@gmail.com wrote:
>>If the goal is newbie adoption, it sounds like there's a much easier
>>solution, based on work that's already been done for us by hard
>>workers.
>
>
> The goal isn't newbie adoption, but if it were, you'd be right, Lispbox
> is good. Regardless, what you are in fact exactly right about is that
> building and maintaining Lispbox is hard work done by hard workers, and
> the guy/folks who does/do it need to pay his/their rent and
> occasionally eat. As has been pointed out, $10,000 isn't much, but it's
> something. I don't mean this in a nasty way, but I'll bet you and the
> rest of the community aren't going to offer the Lispbox guy/folks a
> dime for all his/their hard work, nor will you (that is, the community)
> offer him/them, nor anyone else anything but a warm handshake for the
> work involved in doing what it takes to keep Lisp alive. The open
> source/freeware mentality has broken something deep in the world, and
> it's killing Lisp, and probably lots of other good things.
Well, nothing is going to kill Lisp. What may get killed is people
trying to make money off commodity software, because they will be up
against free versions and it is hard to be better enough to justify
bucks. Especially when some people do not realize how much the rough
edges of "free" stuff costs them.
> Companies
> that build fantastic Lisp platforms can barely stay in business and
> need to scratch for every dime to pay their hard-working engineers,
Agreed. But look at CLisp. That sets the bar $$$ Lisps must clear by a
high enough margin to justify the $$$. And Lispworks sets the bar for
Franz in re runtime licensing. I mention only win32-portable Lisps
because commercial success kinda requires that.
As a Lisp entrepreneur myself I am of course rooting for the commercial
Lisp vendors, but am I supposed to treat them as charity cases and give
them revenue for no reason when I could be using cheaper alternatives
(assuming for the sake of argument that CLisp and LW do not fall into
the "expensive free" category).
Note, btw, that your $10k would be financing another knife in the
$$$Lisp back, since all they have to point to are add-ons like CAPI (LTk
and Cells-Gtk killed that advantage) and AllegroCache. And their
proprietary implementations of sockets and FFI (CFFI killed that) do a
little to lock in developers (not much, tho).
The answer is simple: vertical apps, not tools. It is time for the IT
community to stop creating tools (or trying to make money off them) and
start using them to solve problems in the real world. Paul Graham did
it. Orbitz did it. That is what I have been trying to do for seven years.
All aboard?
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
> Well, nothing is going to kill Lisp.
You're right. It's already dead. (In the sense that it's stopped
evolving-- which is what this was all about to begin with.)
> What may get killed is people
> trying to make money off commodity software, because they will be up
> against free versions and it is hard to be better enough to justify
> bucks. Especially when some people do not realize how much the rough
> edges of "free" stuff costs them.
Amen to that!
> The answer is simple: vertical apps, not tools. It is time for the IT
> community to stop creating tools (or trying to make money off them) and
> start using them to solve problems in the real world. Paul Graham did
> it. Orbitz did it. That is what I have been trying to do for seven years.
>
> All aboard?
I would be, if you were right, but unfortunately, you're not. "Vertical
apps" is too simplistic a way of thinking about it. I work primarily in
biocomputing (which, btw, I do entirely in Lisp!) where freeware
vertical apps abound too, and thus the vertical app manufacturers are
scratching around just like the platform manufacturers in the IT world
are.
It's utterly amazing to me that the folks in the labs that I work with
will drop $50,000 without even thinking about it hard for an HPLC
machine that they'll use, maybe 10 times a year, but I have to beg them
for $1500 to buy the Lisp platform that I use to support every
biocomputing activity in the lab (which is a LOT, and a LOT more work
than the HPLC machine supports.)
The answer isn't vertical or horizontal or anything so simple, and it's
not real or virtual -- what I do for the biologists in my lab is as
real as what that HPLC machine does -- do I get paid for it? No. Can I
get them to pay for software licenses (vertical or horizontal)? No.
It's something like: "Do something that no one can replicate, and that
everyone needs, and charge for it." No one can replicate what an HPLC
machine does, but the biologists <i>believe</i> that they can replicate
what I do for them... with excel or something (which, of course, they
think is free because it <i>appears</i> to be free on the Stanford site
license). In fact, I (and a team of mostly volunteer, but paid when I
get pay them) engineers built them a lovely platform (entirely on Lisp)
with which to do their work... will they help pay for it at all? No.
Why? Well, because it's on the web, and it's just a web thing, and all
web things are just free, aren't they?
The world is all sooooooooooooo f'ed up with respect to software and
money that I just can't believe it...
From: Ken Tilton
Subject: Re: A Gentle proposal: Coup d'etat
Date:
Message-ID: <QSS1g.16$7y4.4@fe11.lga>
········@gmail.com wrote:
>>Well, nothing is going to kill Lisp.
>
>
> You're right. It's already dead. (In the sense that it's stopped
> evolving-- which is what this was all about to begin with.)
Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
almost killed it, but they died a natural and early death because
s-exprs were perfect.
You guys have now been backed into a corner and freely confess that the
only thing missing is libraries that have, with the Internet, recently
become a big deal. And those libraries are not even missing. Every
implementation has them. So really, this is just about it being a little
annoying to develop a portable library, cuz you have to type #+this and
#-that all over the place.
But you cannot even complain about that! Last year Luis rounded out
James's CFFI and filled in one checkbox. This year LTk and Celtk rule
the portable gui roost, tho some prefer Cells-Gtk. Some of us who
Actually Code are talking about re-visiting cl-sockets. And so it goes...
>
>
>>What may get killed is people
>>trying to make money off commodity software, because they will be up
>>against free versions and it is hard to be better enough to justify
>>bucks. Especially when some people do not realize how much the rough
>>edges of "free" stuff costs them.
>
>
> Amen to that!
>
>
>>The answer is simple: vertical apps, not tools. It is time for the IT
>>community to stop creating tools (or trying to make money off them) and
>>start using them to solve problems in the real world. Paul Graham did
>>it. Orbitz did it. That is what I have been trying to do for seven years.
>>
>>All aboard?
>
>
> I would be, if you were right, but unfortunately, you're not. "Vertical
> apps" is too simplistic a way of thinking about it. I work primarily in
> biocomputing (which, btw, I do entirely in Lisp!) where freeware
> vertical apps abound too, and thus the vertical app manufacturers are
> scratching around just like the platform manufacturers in the IT world
> are.
yeah, there are some surprisingly vertical open source apps out there.
But I know of a few people who carved out handy little income streams by
going into (I will make this up so as not to attract more competition)
health club software. Started as an exercise for one health club
owner/friend who was going nuts with their crappy package. And believe
me, all that software out there for dedicated niches like that are crap
compared to what a solid Lisper with half a brain could do.
Lispniks just have to pull themselves away from c.l.l and go to more
cocktail parties or bars and listen to people cry about their systems.
This will usually be your better than average health club owner, because
they are smart enough to know things should be better. Then ask them if
they want a better system and a 5% royalty for a few hours analysis and
to serve as the alpha test site.
>
> It's utterly amazing to me that the folks in the labs that I work with
> will drop $50,000 without even thinking about it hard for an HPLC
> machine that they'll use, maybe 10 times a year, but I have to beg them
> for $1500 to buy the Lisp platform that I use to support every
> biocomputing activity in the lab (which is a LOT, and a LOT more work
> than the HPLC machine supports.)
>
> The answer isn't vertical or horizontal or anything so simple, and it's
> not real or virtual -- what I do for the biologists in my lab is as
> real as what that HPLC machine does -- do I get paid for it? No. Can I
> get them to pay for software licenses (vertical or horizontal)? No.
> It's something like: "Do something that no one can replicate, and that
> everyone needs, and charge for it." No one can replicate what an HPLC
> machine does, but the biologists <i>believe</i> that they can replicate
> what I do for them... with excel or something (which, of course, they
> think is free because it <i>appears</i> to be free on the Stanford site
> license). In fact, I (and a team of mostly volunteer, but paid when I
> get pay them) engineers built them a lovely platform (entirely on Lisp)
> with which to do their work... will they help pay for it at all? No.
> Why? Well, because it's on the web, and it's just a web thing, and all
> web things are just free, aren't they?
>
> The world is all sooooooooooooo f'ed up with respect to software and
> money that I just can't believe it...
>
I like the bit about the FBI using two of the same three vendors for the
next $1b they plan to sink into failed sofwtare. Don't get me started!
But that is why the key is retail sales, not megatools for corporations,
unis, or gov't. I only need to sell to a few percent of any given market
to catch up with Ron; it is OK if 97% of them are too dumb to know their
sofwtare sucks and use it because "everyone else does".
:)
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Ken Tilton wrote:
>
>
> ········@gmail.com wrote:
>
>>> Well, nothing is going to kill Lisp.
>>
>>
>>
>> You're right. It's already dead. (In the sense that it's stopped
>> evolving-- which is what this was all about to begin with.)
>
>
> Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
> almost killed it, but they died a natural and early death because
> s-exprs were perfect.
Dayyum, boy! You still using LISP 1.5? Oh wait, that was released in
1959 or '60. In fact, John McCarthy says he discovered Lisp in 1958.
Now I see how come you have the perfect Lisp -- you've got the imaginary
one that never was. Can you find some way to open-source it, or at
least publish a comprehensive description of the perfect 1956 Lisp? The
world awaits with bated breath.
--Larry
Larry Elmore wrote:
> Ken Tilton wrote:
>
>>
>>········@gmail.com wrote:
>>
>>
>>>>Well, nothing is going to kill Lisp.
>>>
>>>
>>>
>>>You're right. It's already dead. (In the sense that it's stopped
>>>evolving-- which is what this was all about to begin with.)
>>
>>
>>Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
>>almost killed it, but they died a natural and early death because
>>s-exprs were perfect.
>
>
> Dayyum, boy! You still using LISP 1.5? Oh wait, that was released in
> 1959 or '60. In fact, John McCarthy says he discovered Lisp in 1958.
> Now I see how come you have the perfect Lisp -- you've got the imaginary
> one that never was. Can you find some way to open-source it, or at
> least publish a comprehensive description of the perfect 1956 Lisp? The
> world awaits with bated breath.
Here ya go, smart-ass:
> Early thoughts about a language that eventually became Lisp started in 1956 when John McCarthy
> attended the Dartmouth Summer Research Project on Arti cial Intelligence.
> http://www.cs.umbc.edu/331/resources/papers/Evolution-of-Lisp.pdf:
This must be terribly embarrassing for you.
:)
ken
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Ken Tilton wrote:
>
>
> Larry Elmore wrote:
>
>> Ken Tilton wrote:
>>
>>>
>>> ········@gmail.com wrote:
>>>
>>>
>>>>> Well, nothing is going to kill Lisp.
>>>>
>>>>
>>>>
>>>>
>>>> You're right. It's already dead. (In the sense that it's stopped
>>>> evolving-- which is what this was all about to begin with.)
>>>
>>>
>>>
>>> Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
>>> almost killed it, but they died a natural and early death because
>>> s-exprs were perfect.
>>
>>
>>
>> Dayyum, boy! You still using LISP 1.5? Oh wait, that was released in
>> 1959 or '60. In fact, John McCarthy says he discovered Lisp in 1958.
>> Now I see how come you have the perfect Lisp -- you've got the imaginary
>> one that never was. Can you find some way to open-source it, or at
>> least publish a comprehensive description of the perfect 1956 Lisp? The
>> world awaits with bated breath.
>
>
> Here ya go, smart-ass:
>
>> Early thoughts about a language that eventually became Lisp started in
>> 1956 when John McCarthy
>> attended the Dartmouth Summer Research Project on Arti cial Intelligence.
>>
>> http://www.cs.umbc.edu/331/resources/papers/Evolution-of-Lisp.pdf:
>
>
> This must be terribly embarrassing for you.
Not at all, since you apparently don't even comprehend what you quoted.
By the way, you can be one of the proud owners of the world's first
perfect anti-gravity flying car, and not just the 4th or 5th generation
improved model, but the *perfect* model I started having thoughts about
last night as I was drifting off to sleep. Hell, the first prototype
won't even be *started on* for another two years! And it can be yours,
YOURS, *YOURS* for the low, LOW, *LOW* price of $49,995!!! (not
including shipping and handling, tax applicable in some states)
--Larry
Larry Elmore wrote:
> Ken Tilton wrote:
>
>>
>>Larry Elmore wrote:
>>
>>
>>>Ken Tilton wrote:
>>>
>>>
>>>>········@gmail.com wrote:
>>>>
>>>>
>>>>
>>>>>>Well, nothing is going to kill Lisp.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>You're right. It's already dead. (In the sense that it's stopped
>>>>>evolving-- which is what this was all about to begin with.)
>>>>
>>>>
>>>>
>>>>Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
>>>>almost killed it, but they died a natural and early death because
>>>>s-exprs were perfect.
>>>
>>>
>>>
>>>Dayyum, boy! You still using LISP 1.5? Oh wait, that was released in
>>>1959 or '60. In fact, John McCarthy says he discovered Lisp in 1958.
>>>Now I see how come you have the perfect Lisp -- you've got the imaginary
>>>one that never was. Can you find some way to open-source it, or at
>>>least publish a comprehensive description of the perfect 1956 Lisp? The
>>>world awaits with bated breath.
>>
>>
>>Here ya go, smart-ass:
>>
>>
>>>Early thoughts about a language that eventually became Lisp started in
>>>1956 when John McCarthy
>>>attended the Dartmouth Summer Research Project on Artificial Intelligence.
>>>
>>>http://www.cs.umbc.edu/331/resources/papers/Evolution-of-Lisp.pdf:
>>
>>
>>This must be terribly embarrassing for you.
>
>
> Not at all, since you apparently don't even comprehend what you quoted.
Oh, sorry, I thought it meant this:
"This paper concentrates on the development of the basic ideas and
distinguishes two periods - Summer 1956 through Summer 1958 when most of
the key ideas were developed...."
http://coblitz.codeen.org:3125/citeseer.ist.psu.edu/cache/papers/cs/25867/http:zSzzSzwww-formal.stanford.eduzSzjmczSzhistoryzSzlisp.pdf/mccarthy78history.pdf
What ideas? I do not know, but those are the opening words of the
introduction to the "History of Lisp", 2/12/1979, by some guy named John
McCarthy.
Tip #1: Quit while you are behind.
Tip #2: Your attempts at humor are too heavy-handed. ease up on the
throttle a little. You have to leave some work for the audience.
Tip #3: When someone is standing on a chair waving a pint in the air
singing a eulogy to something, be it Lisp or the local football club,
well, no one likes a wise-ass. :)
peace, kenny
--
Cells: http://common-lisp.net/project/cells/
"Have you ever been in a relationship?"
Attorney for Mary Winkler, confessed killer of her
minister husband, when asked if the couple had
marital problems.
Ken Tilton wrote:
>
>
> Larry Elmore wrote:
>
>> Ken Tilton wrote:
>>
>>>
>>> Larry Elmore wrote:
>>>
>>>
>>>> Ken Tilton wrote:
>>>>
>>>>
>>>>> ········@gmail.com wrote:
>>>>>
>>>>>
>>>>>
>>>>>>> Well, nothing is going to kill Lisp.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> You're right. It's already dead. (In the sense that it's stopped
>>>>>> evolving-- which is what this was all about to begin with.)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Lisp does need to evolve. is perfect. It was perfect in 1956. m-exprs
>>>>> almost killed it, but they died a natural and early death because
>>>>> s-exprs were perfect.
>>>>
>>>>
>>>>
>>>>
>>>> Dayyum, boy! You still using LISP 1.5? Oh wait, that was released in
>>>> 1959 or '60. In fact, John McCarthy says he discovered Lisp in 1958.
>>>> Now I see how come you have the perfect Lisp -- you've got the
>>>> imaginary
>>>> one that never was. Can you find some way to open-source it, or at
>>>> least publish a comprehensive description of the perfect 1956 Lisp?
>>>> The
>>>> world awaits with bated breath.
>>>
>>>
>>>
>>> Here ya go, smart-ass:
>>>
>>>
>>>> Early thoughts about a language that eventually became Lisp started in
>>>> 1956 when John McCarthy
>>>> attended the Dartmouth Summer Research Project on Artificial
>>>> Intelligence.
>>>>
>>>> http://www.cs.umbc.edu/331/resources/papers/Evolution-of-Lisp.pdf:
>>>
>>>
>>>
>>> This must be terribly embarrassing for you.
>>
>>
>>
>> Not at all, since you apparently don't even comprehend what you quoted.
>
>
> Oh, sorry, I thought it meant this:
>
> "This paper concentrates on the development of the basic ideas and
> distinguishes two periods - Summer 1956 through Summer 1958 when most of
> the key ideas were developed...."
>
> http://coblitz.codeen.org:3125/citeseer.ist.psu.edu/cache/papers/cs/25867/http:zSzzSzwww-formal.stanford.eduzSzjmczSzhistoryzSzlisp.pdf/mccarthy78history.pdf
>
>
> What ideas? I do not know, but those are the opening words of the
> introduction to the "History of Lisp", 2/12/1979, by some guy named John
> McCarthy.
In other words, the "perfect Lisp" that you use is the one that existed
before any implementation work even started? All I can say is, "Wow, I
wish my meds were that good." :)
--Larry
On 2006-04-21 20:56:18 -0400, Larry Elmore <·········@comcast.net> said:
> All I can say is, "Wow, I
> wish my meds were that good." :)
That's what the pint is for ;^)
Raffael Cavallaro wrote:
> On 2006-04-21 20:56:18 -0400, Larry Elmore <·········@comcast.net> said:
>
>> All I can say is, "Wow, I
>> wish my meds were that good." :)
>
>
> That's what the pint is for ;^)
That explains it then. All I get are these diabetic meds with warnings
not to mix with alcohol. :-(
--Larry
> Of course he relalizes how little it is. He didn't just offer the
cash,
> he challenged others to step up like he did. You don't have to give
> $10,000 -- how about $1000? How about $500? How about $100?
Hey, this is easy. Just get a PO box and...
"My name is Dave Rhodes, err, Common Lisp..."
Title it: "Introduce Misfeatures Fast". "For every appallingly wrong
idea you send out, you are _guaranteeed_ to receive 30,000 badly though
out "improvements". No longer will C++ lord it over you- you too can be
rich in botched semantics if you follow my advice. This program has
remained successful because of the unimpeachable, err, well, because
Larry Wall sits on our board. Please continue its success by CAREFULLY
ADHERING to the instructions."
Send the first copy to EN, just for fun.
Trust me- I've been in prison!
········@gmail.com wrote:
> Tim Bradshaw wrote:
>
> > ... though the original poster has mentioned money, I don't think
> > he realises how little it is ...
>
> Of course he relalizes how little it is. He didn't just offer the cash,
> he challenged others to step up like he did. You don't have to give
> $10,000 -- how about $1000? How about $500? How about $100?
I don't think you realise who ill-conceived I think the idea is.
--tim