I have two versions of CLHS. While I was able to read first version
(about two years old) as a tale, beginning from multilevel contents
and moving further, I can't say the same about new version. Certainly
it contain more (9Mb more) information which is most likely more
consistent, but I found it difficult to find it for beginner.
For example, beginning of the old section about symbols:
Symbols are Lisp data objects that serve several purposes and have
several interesting characteristics. Every object of type symbol has a
name, called its print name.
In the new version I can't find that so easy. Symbols chapter contain
following words:
10.1 Symbol Concepts
The next figure lists some defined names that are applicable to the
property lists of symbols.
...
Figure 10-1. Property list defined names
The next figure lists some defined names that are applicable to the
creation of and inquiry about symbols.
...
Figure 10-2. Symbol creation and inquiry defined names
It's somewhat differ from concepts. True concepts are accessible
through glossary, but the latter is not structured (as it is
glossary) and I believe these concepts are intended .more for experts
than for those who have no idea about symbols.
So I consider these two versions of hyperspec different documents each
with its own value.
On 28 Jul 2001 03:09:15 +1200, Paul Foley <···@below> wrote:
>> I have two versions of CLHS. While I was able to read first version
>> (about two years old) as a tale, beginning from multilevel contents
>> and moving further, I can't say the same about new version.
>
>That would be CLtL2, not the HyperSpec.
You're right. I've been under delusion that html version of CLTL2 is
the hyperspec.
In article <················@news.vtc.ru>,
Arseny Slobodjuck <····@crosswinds.net> wrote:
>On 28 Jul 2001 03:09:15 +1200, Paul Foley <···@below> wrote:
>
>>> I have two versions of CLHS. While I was able to read first version
>>> (about two years old) as a tale, beginning from multilevel contents
>>> and moving further, I can't say the same about new version.
>>
>>That would be CLtL2, not the HyperSpec.
>
>You're right. I've been under delusion that html version of CLTL2 is
>the hyperspec.
I've heard that the HS is more correct than CLtL2, by
virtue of being more recent. However, CLtL2 continues
to be in print, which isn't a surprise considering
GLS's accessible style, but there don't seem to be any
changes/corrections that would make it up-to-date. I
wonder if anyone has listed some marginal notes to go
with CLtL2 so a reader doesn't go astray.
--d
····@goldshoe.gte.com (Dorai Sitaram) writes:
> In article <················@news.vtc.ru>,
> Arseny Slobodjuck <····@crosswinds.net> wrote:
> >On 28 Jul 2001 03:09:15 +1200, Paul Foley <···@below> wrote:
> >
> >>> I have two versions of CLHS. While I was able to read first version
> >>> (about two years old) as a tale, beginning from multilevel contents
> >>> and moving further, I can't say the same about new version.
> >>
> >>That would be CLtL2, not the HyperSpec.
> >
> >You're right. I've been under delusion that html version of CLTL2 is
> >the hyperspec.
>
> I've heard that the HS is more correct than CLtL2, by
> virtue of being more recent. However, CLtL2 continues
> to be in print, which isn't a surprise considering
> GLS's accessible style, but there don't seem to be any
> changes/corrections that would make it up-to-date. I
> wonder if anyone has listed some marginal notes to go
> with CLtL2 so a reader doesn't go astray.
Ah, yes. Essentially adding in some "errata due to new standard."
That would be a _nice_ thing...
--
(reverse (concatenate 'string ··········@" "enworbbc"))
http://vip.hyperusa.com/~cbbrowne/rdbms.html
"What a depressingly stupid machine."
-- Marvin the Paranoid Android
········@hex.net writes:
> ····@goldshoe.gte.com (Dorai Sitaram) writes:
> > In article <················@news.vtc.ru>,
> > Arseny Slobodjuck <····@crosswinds.net> wrote:
> > >On 28 Jul 2001 03:09:15 +1200, Paul Foley <···@below> wrote:
> > >
...
> > I've heard that the HS is more correct than CLtL2, by
> > virtue of being more recent. However, CLtL2 continues
> > to be in print, which isn't a surprise considering
> > GLS's accessible style, but there don't seem to be any
> > changes/corrections that would make it up-to-date. I
> > wonder if anyone has listed some marginal notes to go
> > with CLtL2 so a reader doesn't go astray.
>
> Ah, yes. Essentially adding in some "errata due to new standard."
> That would be a _nice_ thing...
(rant warning)
CLTL2 was not EVER a standard.
The parent of ANSI CL is CLTL, not CLTL2. CLTL2 is a sibling of ANSI CL,
not a parent.
CLTL and ANSI CL were committee products. They were reviewed by committees
and approved by committees both in terms of content and presentation.
CLTL2 was a personal project. It was not reviewed by nor approved by any
committee. It was a snapshot of work in progress taken at a random time
by an individual without the encouragement (and with some DIScouragement)
of the committee. That's why it contains things that were half-baked,
later retracted, later amended, etc.
No vendor was intended to adopt this document as a reference work.
Some tried. That complicated matters.
Btw, the HyperSpec is not officially ANSI CL either, however, it IS
officially (i.e., in a legally authorized-by-ANSI sense) a derivative
work of the document that is the standard. (The similar document
offered by Franz is, I believe, not a derivative work of ANSI CL, but
the sense in which it is not is of very little importance, since it's
a derivative of another document which had only non-technical changes
made before it became the ANSI CL standard.) I mention all of this
just to underscore the fact that, strictly, the ANSI standard is a
plaintext work available only from ANSI, and to help people understand
the lineage.
I II III
CLTL ---+----------> final draft CL spec ---+---> ANSI CL --> CLHS
| |
+--> CLTL2 +-----------------> Franz CL ref
Note that technical changes continued to be made in intervals I and II,
but not in interval III or later.
I feel a need to reiterate this every time anyone talks about ANSI CL or
CLHS as a "successor" to CLTL2. It simply is not, other than in the trivial
sense of time-ordering.
Kent M Pitman <······@world.std.com> writes:
> ········@hex.net writes:
> > ····@goldshoe.gte.com (Dorai Sitaram) writes:
> > > In article <················@news.vtc.ru>,
> > > Arseny Slobodjuck <····@crosswinds.net> wrote:
> > > >On 28 Jul 2001 03:09:15 +1200, Paul Foley <···@below> wrote:
> > > >
> ....
> > > I've heard that the HS is more correct than CLtL2, by
> > > virtue of being more recent. However, CLtL2 continues
> > > to be in print, which isn't a surprise considering
> > > GLS's accessible style, but there don't seem to be any
> > > changes/corrections that would make it up-to-date. I
> > > wonder if anyone has listed some marginal notes to go
> > > with CLtL2 so a reader doesn't go astray.
> >
> > Ah, yes. Essentially adding in some "errata due to new standard."
> > That would be a _nice_ thing...
>
> (rant warning)
> CLTL2 was not EVER a standard.
The preface nicely states that.
> I feel a need to reiterate this every time anyone talks about ANSI
> CL or CLHS as a "successor" to CLTL2. It simply is not, other than
> in the trivial sense of time-ordering.
I don't think anybody talked about any such "formal succession."
CLtL2 was the nearest thing to documentation of Common Lisp until the
ANSI standard and/or CLHS came along; while not official, it was, and
even still is, quite useful, irrespective of any formal status.
CLtL2 is, in particular, exceedingly valuable for having a sizable
number of examples, that being something that was out of scope for the
ANSI standard.
The _point_ is that it would be nice if CLtL2 were further annotated
to better indicate places where ANSI have impact, whether this is
because it "became inaccurate because the standards changed," or, for
less charitable viewpoints, "more crucially wrong than it previously
was."
--
(concatenate 'string "aa454" ·@freenet.carleton.ca")
http://vip.hyperusa.com/~cbbrowne/unix.html
You know better than to trust a strange computer.
········@hex.net writes:
> The _point_ is that it would be nice if CLtL2 were further annotated
> to better indicate places where ANSI have impact, whether this is
> because it "became inaccurate because the standards changed," or, for
> less charitable viewpoints, "more crucially wrong than it previously
> was."
That would make it a CLtL3. If someone were to do that, why not just
document the actual standard, instead of giving errata?
--
Duane Rettig Franz Inc. http://www.franz.com/ (www)
1995 University Ave Suite 275 Berkeley, CA 94704
Phone: (510) 548-3600; FAX: (510) 548-8253 ·····@Franz.COM (internet)
Duane Rettig <·····@franz.com> writes:
> ········@hex.net writes:
>
> > The _point_ is that it would be nice if CLtL2 were further annotated
> > to better indicate places where ANSI have impact, whether this is
> > because it "became inaccurate because the standards changed," or, for
> > less charitable viewpoints, "more crucially wrong than it previously
> > was."
>
> That would make it a CLtL3. If someone were to do that, why not just
> document the actual standard, instead of giving errata?
So elegantly put.
The only answers I can think of are:
* they want to upgrade a CLTL2-compliant body of code
* they have only academic interest and don't understand the value of
the time it would require out of people who have the relevant
information to do this.
Lots of things in this world "would be nice". I call that the "one
place predicate" that deludes people into thinking they can or should
get something that really they should not. Politicians love this
question. "What do you think about the environment, Mr. Senator?"
"I'm for it." "Schools?" "For it." "Lower taxes?" "For it." "Better
defense?" "For it." No choice to be made there. "If lower taxes
meant poorer schools?" "Ha, ha. Very interesting question. Next..."
That's a "two place predicate" slipping in. Forcing a choice.
Let's put it this way to help us understand how "nice" it is to have
this: Anyone who wants to hire me at $100/hr (probably less than what
Steele would charge) to write up the differences can send me private
email to arrange such a contract. It could probably be whipped out in
a "mere" week or two ($4K-$8K, probably, but don't ask me to give you
a guaranteed bound or I'll be forced to say three or four weeks just
to be sure I'm not cutting it too fine), so if you don't mind ante'ing
up a few thousand dollars (feel free to pass the hat, but don't ask me
to), absolutely let's solve this "important" problem.
If no one approaches me on this, then I, at least, will take the
silence to mean it doesn't, after all, "need" to be done. Then again,
maybe I'm just the wrong person to do it--that would be ok, too. :-)
Yes, it would be "nice" in some trivial way to have this. But so many
other things would be "nice" that I find it irrelevant to say this.
The community needs to focus on things that "would matter", and money is
an easy way to find those things.
Add this to my list of reasons that "free software" is not my cup of tea.
It encourages a system in which you don't have a metric for comparing
the value of your use of time on one activity with the value of your use
of time on another. (I assume that's the intent, since some people would
not want the mere fact that something has cost to keep them from getting
ahold of it, but a side-effect is that sometimes people who could be doing
something more important with their lives are fooled into doing something
less important because they don't make a regular practice of measuring
the importance of what they are doing with this common yardstick. That's
not to say that it's the only yardstick by which any human endeavor should
be measured--only that hiding from the measure is as bad as adhering solely
to the measure.)
On Mon, 30 Jul 2001, Kent M Pitman wrote:
> Duane Rettig <·····@franz.com> writes:
> > ········@hex.net writes:
> >
> > > The _point_ is that it would be nice if CLtL2 were further annotated
> > > to better indicate places where ANSI have impact, whether this is
> > > because it "became inaccurate because the standards changed," or, for
> > > less charitable viewpoints, "more crucially wrong than it previously
> > > was."
> >
> > That would make it a CLtL3. If someone were to do that, why not just
> > document the actual standard, instead of giving errata?
>
> So elegantly put.
>
> The only answers I can think of are:
>
> * they want to upgrade a CLTL2-compliant body of code
>
> * they have only academic interest and don't understand the value of
> the time it would require out of people who have the relevant
> information to do this.
* they are using an implementation that is not quite ANSI compliant and
need some clues as to why that might be so and possible workarounds;
* the historical record, while possibly only of academic interest, gives
them warm fuzzies.
At this point, an "errata to the errata" style would probably be too
unwieldy; it'd be much less confusing to have the main text reflect the
current standard, with notes describing changes from CLTL and CLTL2. Not
that it's ever going to happen.
> [ kmp's generous offer]
> If no one approaches me on this, then I, at least, will take the
> silence to mean it doesn't, after all, "need" to be done. Then again,
> maybe I'm just the wrong person to do it--that would be ok, too. :-)
No offense, but I would think that you're the wrong person for the job, if
only for your not-so-subtle hints that you disapproved of CLTL2 to begin
with.
> Yes, it would be "nice" in some trivial way to have this. But so many
> other things would be "nice" that I find it irrelevant to say this.
> The community needs to focus on things that "would matter", and money is
> an easy way to find those things.
I believe people think it's especially "nice" because they have a soft
spot for CLTL and CLTL2. CLTL2 was well written, had interesting example
code, some clever GLS jokes, and described a language that was (and
remains) far and away better than anything else. It was a very optimistic
document for a time of encroaching darkness.
Now, at the beginning of a new millenium, there may be less reason for an
updated CLTL2 because other current resources fulfill its functions
better: we have good introductions to the language (Graham's books) and
authoritative references (The HyperSpec, the crown jewel of the Lisp
world). However, if people think an updated CLTL2 would be nice, I
suggest below a mechanism to make it happen.
> Add this to my list of reasons that "free software" is not my cup of tea.
> It encourages a system in which you don't have a metric for comparing
> the value of your use of time on one activity with the value of your use
> of time on another. (I assume that's the intent, since some people would
> not want the mere fact that something has cost to keep them from getting
> ahold of it, but a side-effect is that sometimes people who could be doing
> something more important with their lives are fooled into doing something
> less important because they don't make a regular practice of measuring
> the importance of what they are doing with this common yardstick. That's
> not to say that it's the only yardstick by which any human endeavor should
> be measured--only that hiding from the measure is as bad as adhering solely
> to the measure.)
One of the reasons that free software is my cup of tea is its mantra of
"many hands make light work." In this case, both CLTL2 and the HyperSpec
exist on the web. An interesting programming exercise for some aspiring
Lisp hotshot would be to adapt the Cliki software to allow annotations of
the underlaying web pages via a proxy. This isn't a new idea, of course;
several dot coms have gone broke with this very concept :) Anyway, one
could imagine a Talmudic layering of commentary on the HyperSpec and a
CLTL2 updated by the community.
Tim
>>>>> "TM" == Tim Moore <·····@herschel.bricoworks.com> writes:
[...]
TM> I believe people think it's especially "nice" because they
TM> have a soft spot for CLTL and CLTL2. CLTL2 was well written,
TM> had interesting example code, some clever GLS jokes, and
TM> described a language that was (and remains) far and away
TM> better than anything else. It was a very optimistic document
TM> for a time of encroaching darkness. [...]
While I share the mush about the CLTL family, I have trouble understanding
the rest of this paragraph. Do you mean to say that the ANSI CL is worse
than the language described the CLTL2?
cheers,
BM
On Mon, 30 Jul 2001, Bulent Murtezaoglu wrote:
> >>>>> "TM" == Tim Moore <·····@herschel.bricoworks.com> writes:
> [...]
> TM> I believe people think it's especially "nice" because they
> TM> have a soft spot for CLTL and CLTL2. CLTL2 was well written,
> TM> had interesting example code, some clever GLS jokes, and
> TM> described a language that was (and remains) far and away
> TM> better than anything else. It was a very optimistic document
> TM> for a time of encroaching darkness. [...]
>
> While I share the mush about the CLTL family, I have trouble understanding
> the rest of this paragraph. Do you mean to say that the ANSI CL is worse
> than the language described the CLTL2?
Oh no, not at all. By "encroaching darkness" I meant the AI Winter;
increasing desperation of Lisp vendors; fatigue within X3J13, especially
of representitives who were users, etc. At the time I thought it was a
pretty cool statement that someone was willing to publish a 1000+ page
book on Common Lisp.
I realize that my comment could be construed to mean "described a
language that was better than anything else, including ANSI CL." I meant
"described a language, the evolving ANSI CL, that was better than CLTL
Common Lisp, C, C++, Modula 2, etc."
Tim
Tim Moore <·····@herschel.bricoworks.com> writes:
> On Mon, 30 Jul 2001, Bulent Murtezaoglu wrote:
>
> Oh no, not at all. By "encroaching darkness" I meant the AI Winter;
FWIW, I took this to be your meaning and didn't assume you were
casting aspersions on the language. I think to those of us who went
through AI Winter, perhaps, it was easier to remember the "bigness" of
this event, and so easier to understand the context of your sentence.
Maybe to those not present or intensely involved at the time, the
reference was more obscure.
Tim Moore <·····@herschel.bricoworks.com> writes:
> No offense, but I would think that you're the wrong person for the job, if
> only for your not-so-subtle hints that you disapproved of CLTL2 to begin
> with.
There might be a lot of reasons why I might be the wrong person, but
this would not be one. There was likewise some concern when I
volunteered as Editor of ANSI CL that I might not be able to separate
my strong technical passions for where I wanted the language to go
from the objectivity needed to be Editor (often refered to as "neutral
scribe"); I hope I ultimately put such concerns to rest by just being
careful about when I was being a technical advocate and allowed to
show passion and when I was being neutral and required to do as
committee vote says. This is an important skill I'm proud to claim
and that I wish everyone were actively trained in, as it would facilitate
technical discussion under fixed sets of axioms a great deal in some cases,
avoiding useless discussions on issues outside of agreed-upon parameters.
You can dislike me or my technical or political positions for all
kinds of reasons, and I will try to take it in stride, but I took this
quoted remark above as a personal slight, and utterly without
foundation. I am a professional and as capable as any among us of
being objective when a job calls for it. An apology is not required, but
I wanted to speak to the matter myself since it is, as I said, a matter
of great personal pride.
On Mon, 30 Jul 2001, Kent M Pitman wrote:
> Tim Moore <·····@herschel.bricoworks.com> writes:
>
> > No offense, but I would think that you're the wrong person for the job, if
> > only for your not-so-subtle hints that you disapproved of CLTL2 to begin
> > with.
>
> There might be a lot of reasons why I might be the wrong person, but
> this would not be one. There was likewise some concern when I
> volunteered as Editor of ANSI CL that I might not be able to separate
> my strong technical passions for where I wanted the language to go
> from the objectivity needed to be Editor (often refered to as "neutral
> scribe"); I hope I ultimately put such concerns to rest by just being
> careful about when I was being a technical advocate and allowed to
> show passion and when I was being neutral and required to do as
> committee vote says. This is an important skill I'm proud to claim
> and that I wish everyone were actively trained in, as it would facilitate
> technical discussion under fixed sets of axioms a great deal in some cases,
> avoiding useless discussions on issues outside of agreed-upon parameters.
I have absolutely no disagreement with that.
> You can dislike me or my technical or political positions for all
> kinds of reasons, and I will try to take it in stride, but I took this
> quoted remark above as a personal slight, and utterly without
> foundation. I am a professional and as capable as any among us of
> being objective when a job calls for it. An apology is not required, but
> I wanted to speak to the matter myself since it is, as I said, a matter
> of great personal pride.
I certainly do apologize. I agree with almost all of your technical
positions and very few of your political positions, but I sincerly did
not mean any kind of personal slight. I did not mean that one should have
any concern with the quality of work you would do in updating CLTL2, more
that since you had opposed the existence of CLTL2 (and I still see no
protests to the contrary from you on that score) that you would be doing
work that you fundamentally don't like. Perhaps that's a false assumption
on my part, but if not, I would think you'd have something better to do.
Tim
Kent M Pitman <······@world.std.com> writes:
> Add this to my list of reasons that "free software" is not my cup of tea.
> It encourages a system in which you don't have a metric for comparing
> the value of your use of time on one activity with the value of your use
> of time on another. (I assume that's the intent, since some people would
> not want the mere fact that something has cost to keep them from getting
> ahold of it, but a side-effect is that sometimes people who could be doing
> something more important with their lives are fooled into doing something
> less important because they don't make a regular practice of measuring
> the importance of what they are doing with this common yardstick.
Given that a lot of open source work is done by programmers/users
"scratching an itch", it gets evaluated against the value of the
programmers' time. If I find some behaviour of <open source program
foo> is missing/wrong/bothering me, I can investigate how long it
would take me to correct it. Once I've done this work, I'd rather have
it incorporated in the official version since keeping it as a separate
patch that I have to keep in sync with new releases adds work for me.
--
Lieven Marchand <···@wyrd.be>
You can drag any rat out of the sewer and teach it to get some work done in
Perl, but you cannot teach it serious programming. Erik Naggum
Lieven Marchand <···@wyrd.be> writes:
> Kent M Pitman <······@world.std.com> writes:
>
> > Add this to my list of reasons that "free software" is not my cup of tea.
> > It encourages a system in which you don't have a metric for comparing
> > the value of your use of time on one activity with the value of your use
> > of time on another. (I assume that's the intent, since some people would
> > not want the mere fact that something has cost to keep them from getting
> > ahold of it, but a side-effect is that sometimes people who could be doing
> > something more important with their lives are fooled into doing something
> > less important because they don't make a regular practice of measuring
> > the importance of what they are doing with this common yardstick.
>
> Given that a lot of open source work is done by programmers/users
> "scratching an itch", it gets evaluated against the value of the
> programmers' time. If I find some behaviour of <open source program
> foo> is missing/wrong/bothering me, I can investigate how long it
> would take me to correct it. Once I've done this work, I'd rather have
> it incorporated in the official version since keeping it as a separate
> patch that I have to keep in sync with new releases adds work for me.
Once again let me say that I am not trying to utterly condem the open
source approach, only to add balance to the many opposing chants I
hear.
Notwithstanding any of my rhetoric on the issue, I've contributed gobs
of openish and freeish software in my time. But my point is that in
retrospect I consider some of that (not all of it) wasted. And not
always because I could have made money, which was the point of this
more recent remark of mine. Rather, sometimes because now
understanding the value of "time", I can more easily see that the
numerous hours spent in "idle foolishness", whether intended for "the
good of mankind" or "just to scratch an itch" could have been spent in
other endeavors that I value more. Money is not the only metric by
which I do such comparisons; in some cases I might say the metric
should be time spent with a loved one. But the point is not to hold
up money or love as the ultimate metric; only to say that thinking
about the value of your time, the one true valuable commodity you will
ever own, rather than just frittering it away as if it had no value,
is important.
In article <···············@world.std.com>,
Kent M Pitman <······@world.std.com> wrote:
>Lieven Marchand <···@wyrd.be> writes:
>
>> Kent M Pitman <······@world.std.com> writes:
>>
>> > Add this to my list of reasons that "free software" is not my cup of tea.
>> > It encourages a system in which you don't have a metric for comparing
>> > the value of your use of time on one activity with the value of your use
>> > of time on another. (I assume that's the intent, since some people would
>> > not want the mere fact that something has cost to keep them from getting
>> > ahold of it, but a side-effect is that sometimes people who could be doing
>> > something more important with their lives are fooled into doing something
>> > less important because they don't make a regular practice of measuring
>> > the importance of what they are doing with this common yardstick.
>>
>> Given that a lot of open source work is done by programmers/users
>> "scratching an itch", it gets evaluated against the value of the
>> programmers' time. If I find some behaviour of <open source program
>> foo> is missing/wrong/bothering me, I can investigate how long it
>> would take me to correct it. Once I've done this work, I'd rather have
>> it incorporated in the official version since keeping it as a separate
>> patch that I have to keep in sync with new releases adds work for me.
>
>Once again let me say that I am not trying to utterly condem the open
>source approach, only to add balance to the many opposing chants I
>hear.
>
>Notwithstanding any of my rhetoric on the issue, I've contributed gobs
>of openish and freeish software in my time. But my point is that in
>retrospect I consider some of that (not all of it) wasted. And not
>always because I could have made money, which was the point of this
>more recent remark of mine. Rather, sometimes because now
>understanding the value of "time", I can more easily see that the
>numerous hours spent in "idle foolishness", whether intended for "the
>good of mankind" or "just to scratch an itch" could have been spent in
>other endeavors that I value more. Money is not the only metric by
>which I do such comparisons; in some cases I might say the metric
>should be time spent with a loved one. But the point is not to hold
>up money or love as the ultimate metric; only to say that thinking
>about the value of your time, the one true valuable commodity you will
>ever own, rather than just frittering it away as if it had no value,
>is important.
Smaller, less frequent posts help. ;-)
--d
In article <···············@world.std.com>,
Kent M Pitman <······@world.std.com> wrote:
>Duane Rettig <·····@franz.com> writes:
>> That would make it a CLtL3. If someone were to do that, why not just
>> document the actual standard, instead of giving errata?
>
>So elegantly put.
>
>The only answers I can think of are:
>
> * they want to upgrade a CLTL2-compliant body of code
Presumably CLTL3 would be notated with change bars, just as CLTL2 was. So
it would provide the same type of information useful for upgrading old code
as a list of errata would.
> * they have only academic interest and don't understand the value of
> the time it would require out of people who have the relevant
> information to do this.
I suspect that most of the work would be involved in finding all the
errata, and merging it into the text of CLTL would be relatively easy.
But we'd have to find someone willing to do all this editorial work. GLS
has since moved on to other things (does anyone know if he's still at Sun
working on Java?).
--
Barry Margolin, ······@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
On Tue, 31 Jul 2001 19:00:31 GMT, Barry Margolin <······@genuity.net>
wrote:
>In article <···············@world.std.com>,
>Kent M Pitman <······@world.std.com> wrote:
>>
>> * they have only academic interest and don't understand the value of
>> the time it would require out of people who have the relevant
>> information to do this.
>
>I suspect that most of the work would be involved in finding all the
>errata, and merging it into the text of CLTL would be relatively easy.
>
>But we'd have to find someone willing to do all this editorial work. GLS
>has since moved on to other things (does anyone know if he's still at Sun
>working on Java?).
>
When you say "we'd have to find someone willing to do all this editorial
work", who are the "we" you have in mind?
Lars
In article <················@news1.telia.com>,
Lars Lundback <···············@telia.com> wrote:
>On Tue, 31 Jul 2001 19:00:31 GMT, Barry Margolin <······@genuity.net>
>wrote:
>>But we'd have to find someone willing to do all this editorial work. GLS
>>has since moved on to other things (does anyone know if he's still at Sun
>>working on Java?).
>>
>
>When you say "we'd have to find someone willing to do all this editorial
>work", who are the "we" you have in mind?
The people in the Lisp community who would like to see a new edition of
CLTL.
--
Barry Margolin, ······@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
On Thu, 02 Aug 2001 00:44:46 GMT, Barry Margolin <······@genuity.net>
wrote:
>
>The people in the Lisp community who would like to see a new edition of
>CLTL.
>
I thought you meant something like that, but I wanted to be sure.
Kent has already offered his services and since I have left my employer
and am looking for job opportunities, so could I, for an appropriate fee
of course. Finding people to do this job is not difficult and the
"community" need not look very hard.
But can the "community" pay for the work to be done? I would be happy to
send a small contribution, say $10, but how is this organized? Who
collects the signs the check?
Strictly, where and what is the Common Lisp community? Are there any
officers representing it? Your name in the ALU Initial Board of
Directors list, but that document (from the ALU web) is 10 years old by
now, and the status of ALU is unclear. Perhaps you can illuminate?
Lars
In article <················@news1.telia.com>,
Lars Lundback <···············@telia.com> wrote:
>But can the "community" pay for the work to be done? I would be happy to
>send a small contribution, say $10, but how is this organized? Who
>collects the signs the check?
I would expect that the authors would expect to be paid royalties from the
publisher, just like GLS was.
>Strictly, where and what is the Common Lisp community? Are there any
>officers representing it? Your name in the ALU Initial Board of
>Directors list, but that document (from the ALU web) is 10 years old by
>now, and the status of ALU is unclear. Perhaps you can illuminate?
My only involvement in Common Lisp these days is reading and posting to
this newsgroup. I don't use Lisp in my work (except in my .emacs file), I
haven't been to a Lisp-related conference in years, and I'm not a member of
J13.
The "Common Lisp community" is not an official body, it's just people who
are enthusiastic about CL. It's like the "bicycling community", which
consists of everyone who happens to ride bikes.
--
Barry Margolin, ······@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
Barry Margolin <······@genuity.net> wrote:
: I haven't been to a Lisp-related conference in years,
November 98.
Because it was right around the corner?
I saw your (very) long hair if I remember that correctly.
In article <············@fstgss02.tu-graz.ac.at>,
Reini Urban <······@x-ray.at> wrote:
>Barry Margolin <······@genuity.net> wrote:
>: I haven't been to a Lisp-related conference in years,
>
>November 98.
> Because it was right around the corner?
>I saw your (very) long hair if I remember that correctly.
Wasn't me. I've never had very long hair. It was shoulder-length when I
was a preteen, but then I joined the Civil Air Patrol and had to get an Air
Force cut, and I've worn it short ever since.
--
Barry Margolin, ······@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
········@hex.net writes:
> The _point_ is that it would be nice if CLtL2 were further annotated
> to better indicate places where ANSI have impact, whether this is
> because it "became inaccurate because the standards changed," or, for
> less charitable viewpoints, "more crucially wrong than it previously
> was."
The point is that CLTL2 *is* so annotated. Everything it reports on
was ANSI work in progress and was true as to the date that CLTL2 was
published. But since Steele did not synchronize that date with us,
and since he exposed all progress (even tentative things that were to
be reversed), think of it as an exercise in the "open source" problem
discussed in another thread where people insisted, as I suggested they
naturally do, on inferring more than the preface told them to do.
You'd need a CLTL3, but Steele was tired and wanted out of that
business; he told me he was happy to see ANSI CL published to keep him
from having to do one. And the committee wanted no CLTL2 in the first
place. So it falls to the community, not to the writers of CLTL2 nor
ANSI CL to do whatever "notes" they want about the differences.
- - - - -
On a related note, and I've told this story before, but it bears
repeating, we had an X3J13 meeting after CLTL2 was published at which
we proposed some change (the renaming of the setf expander functions?)
that would mean renaming some symbol referred to by CLTL2. Someone
present fussed, saying "But then I can't use CLTL2 as a reference any
more." Steele and I were both shocked, since we'd already made other
changes to the semantics of the language since CLTL2 even though not
to symbol names. "Semantics don't matter that much," the person said.
"Then we had better vote for the symbol name changes!" Steele and I
agreed. And I think the vote that was pending on the name changes
went through. People will cling to wrong documents for odd reasons.
(I claim because of the expense of copying wetware. Easier to annotate
a prior tree of knowledge than build a new one.)
On Mon, 30 Jul 2001, Kent M Pitman wrote:
> On a related note, and I've told this story before, but it bears
> repeating, we had an X3J13 meeting after CLTL2 was published at which
> we proposed some change (the renaming of the setf expander functions?)
> that would mean renaming some symbol referred to by CLTL2. Someone
> present fussed, saying "But then I can't use CLTL2 as a reference any
> more." Steele and I were both shocked, since we'd already made other
> changes to the semantics of the language since CLTL2 even though not
> to symbol names. "Semantics don't matter that much," the person said.
> "Then we had better vote for the symbol name changes!" Steele and I
> agreed. And I think the vote that was pending on the name changes
> went through. People will cling to wrong documents for odd reasons.
> (I claim because of the expense of copying wetware. Easier to annotate
> a prior tree of knowledge than build a new one.)
That person was a butthead (and I don't think it was me ;).
Tim