Satisfice?
Spehro Pefhany
used the word "satisfice". That word is not in AH4 or M-W, and Google
yields a paltry 9 hits. It seems to mean something like "first reasonable
choice", as opposed to, say, "optimal choice".
Anyone else run into this word? It's not without usefulness.
Best regards,
--
Spehro Pefhany --"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
David Squire
> I'm reviewing a technical book proposal for a publisher and the author has
> used the word "satisfice". That word is not in AH4 or M-W, and Google
> yields a paltry 9 hits. It seems to mean something like "first reasonable
> choice", as opposed to, say, "optimal choice".
>
> Anyone else run into this word? It's not without usefulness.
Could I see it in a sentence?
D.
Spehro Pefhany
> Could I see it in a sentence?
Certainly: It's necessary to satisfice on the choice of an apartment.
Note: There seem to be two basic definitions (even among the few
references). One is "first reasonable choice" and the other is "a choice
that is (probably) less than perfect, chosen for the moment".
David Squire
> The renowned David Squire <David....@csse.monash.edu.au> wrote:
>
> > Could I see it in a sentence?
>
> Certainly: It's necessary to satisfice on the choice of an apartment.
>
> Note: There seem to be two basic definitions (even among the few
> references). One is "first reasonable choice" and the other is "a choice
> that is (probably) less than perfect, chosen for the moment".
The OED has no problem with it as an intransitive verb, with the meaning you
suggest:
satisfice, -fise, v.
[Alteration of SATISFY (influenced by L. satisfacre).]
1. trans. = SATISFY v. Obs. exc. north. (see E.D.D.).
1561 J. DAUS tr. Bullinger on Apoc. (1573) 168b, That their founders
were nourished by suckyng of a wolfe: so haue all that people wolues
mindes, neuer satisfised with bloud, euer greedy of dominion and
hungryng after riches. 1597 in Feuillerat Revels Q. Eliz. (1908) 417
The other officers will nott be satisficed. 1721 KELLY Scot. Prov. 325
Satisfic'd, that is, satisfied.
2. intr. To decide on and pursue a course of action that will
satisfy the minimum requirements necessary to achieve a particular
goal. Hence satisficer; satisficing ppl. a. and vbl. n.
1956 H. SIMON in Psychol. Rev. LXIII. 129/2 Evidently, organisms
adapt well enough to `satisfice'; they do not, in general, `optimize'.
Ibid. 136/1 A `satisficing' path, a path that will permit satisfaction
at some specified level of all its needs. 1957 Models of Man IV. 205
The key..appeared to lie in substituting the goal of satisficing, of
finding a good enough move, for the goal of minimaxing, of finding the
best move. 1958 MARCH & SIMON Organizations vi. 141 To optimize
requires processes several orders of magnitude more complex than those
required to satisfice. 1963 G. P. E. CLARKSON in A. R. Oxenfeldt
Models of Markets II. 340 Two important innovations..have occurred...
The first of these is the modified concept of rational behavior known
as `satisficing' ... Important changes in the theory of the firm have
been brought about by the introduction of the satisficing concept of
behavior. 1967 H. SIMON in N. Rescher Logic of Decision & Action i. 19
It is easy to see how GPS can be made into a satisficer. 1973 N.Y.
Times 11 Feb. III. 1/2 Big business executives don't really try to
maximize profits but `satisfice'that is, they try to make enough
profit to keep stockholders and boards of directors happy without
bringing the wrath of government regulators, consumer groups or
business competitors down on them. 1977 P. N. KHANDWALLA Design of
Organizations xi. 404 To the seat-of-the-pants `satisficer',
scientific analysis may be acceptable in dealing with relatively
trivial problems. 1977 JANIS & MANN Decision Making ii. 32 A much more
serious flaw of this complex form of satisficing lies in its failure
to ensure that the alternatives retained are..superior to those
eliminated.
On the other hand, will readers understand it? If it is to be used frequently,
it might be worth defining it the first time.
Cheers,
D.
Douwe Egbert
"Spehro Pefhany" <sp...@interlog.com> wrote in message
news:ar348.11365$eL....@news1.bloor.is...
> I'm reviewing a technical book proposal for a publisher and the author has
> used the word "satisfice". That word is not in AH4 or M-W, and Google
> yields a paltry 9 hits. It seems to mean something like "first reasonable
> choice", as opposed to, say, "optimal choice".
>
> Anyone else run into this word? It's not without usefulness.
In what seems to be an artificially Latinate sort
of way, yes, I agree that it is not for nothing that
you say that. Yet it is not totally without uselessness,
either, for most of us.
--
D E
Pat Durkin
> The renowned David Squire <David....@csse.monash.edu.au> wrote:
>
> > Could I see it in a sentence?
>
> Certainly: It's necessary to satisfice on the choice of an
apartment.
>
> Note: There seem to be two basic definitions (even among the few
> references). One is "first reasonable choice" and the other is "a
choice
> that is (probably) less than perfect, chosen for the moment".
OneLook ( http://www.onelook.com/ )finds:
Forthrights Phrontistery - Dictionary of Obscure Words -includes
definition (English) (159KB, indexed 13Jan2002)
Thanks to Stephen Chrisomalis
satisfice to aim for or achieve that which will suffice
Those definitions are nouns, while your example uses the word as a
verb.
In your example, I wonder if "compromise", or "settle" would be a
better choice. The usage simply does not seem correct.
Actually, satisfice as a noun calls to mind artifice as a noun. (So
what if my source treats it like a verb also.)
An interesting challenge. Thank you.
Judy
> I'm reviewing a technical book proposal for a publisher and the author has
> used the word "satisfice". That word is not in AH4 or M-W, and Google
> yields a paltry 9 hits. It seems to mean something like "first reasonable
> choice", as opposed to, say, "optimal choice".
>
> Anyone else run into this word? It's not without usefulness.
>
> Best regards,
I've never hear of "satisfice". The words that do come to mind are
"satisfied" and "satisfactory".
As in it's not the optimal choice but something one can live with.
Judy
Spehro Pefhany
> The OED has no problem with it as an intransitive verb, with the meaning you
> suggest:
<snipped>
Thanks!
> On the other hand, will readers understand it? If it is to be used frequently,
> it might be worth defining it the first time.
Sure. It might be worth popularizing despite its "Obs." status- it doesn't
have the negative connotations of "settle".
(I've been trying to come up with better ways of getting the concept
across of completing "most" of what is required on-time or early. A
complementary concept to the Japanese Kaizen (continous improvement)
principle.)
David Squire
> The renowned David Squire <David....@csse.monash.edu.au> wrote:
>
> > The OED has no problem with it as an intransitive verb, with the meaning you
> > suggest:
>
> <snipped>
>
> Thanks!
>
> > On the other hand, will readers understand it? If it is to be used frequently,
> > it might be worth defining it the first time.
>
> Sure. It might be worth popularizing despite its "Obs." status- it doesn't
> have the negative connotations of "settle".
AFAIR, The "Obs" was only for the transitive form.
> (I've been trying to come up with better ways of getting the concept
> across of completing "most" of what is required on-time or early.
Winging it? :)
Martin Ambuhl
>
> I'm reviewing a technical book proposal for a publisher and the author has
> used the word "satisfice". That word is not in AH4 or M-W, and Google
> yields a paltry 9 hits. It seems to mean something like "first reasonable
> choice", as opposed to, say, "optimal choice".
>
> Anyone else run into this word? It's not without usefulness.
This term is now usually associated with Herb Simon (Economics Nobel
Prize, 1978) who did a lot of work in this area (and many others,
including text analysis). The meaning "decide on and pursue a cource of
action that will satisfy the minimum requirements necessary to achieve a
particular goal" is his, but the word itself as a synonym for satisfy is
at least 450 years old. It is also spelt "satisfise."
The term is useful because information has costs. Most attempts to find
"optimal" solutions are counterproductive; the costs of obtaining and
interpreting the information required almost always swampt any gains.
--
Carius est nobis flagellari p doctrina quam nescire.
[leofre ys us beon beswungen for lare thaenne hit ne cunnan.]
- MS Cotton Tiberius A, xv, fol. 60v (British Library)
William Fulmor
>news:ar348.11365$eL....@news1.bloor.is...
>> I'm reviewing a technical book proposal for a publisher and the author has
>> used the word "satisfice". That word is not in AH4 or M-W, and Google
>> yields a paltry 9 hits. It seems to mean something like "first reasonable
>> choice", as opposed to, say, "optimal choice".
>>
>> Anyone else run into this word? It's not without usefulness.
IIRC, I first encountered 'satisficers' or 'satisficing' in the late
60s or early 70s in the context of economics or business management:
eg: 'American managers are satisficers, not maximizers'.
[good enough for government work]
I have no idea of its source.
Bill
Spehro Pefhany
> This term is now usually associated with Herb Simon (Economics Nobel
> Prize, 1978) who did a lot of work in this area (and many others,
> including text analysis). The meaning "decide on and pursue a cource of
> action that will satisfy the minimum requirements necessary to achieve a
> particular goal" is his, but the word itself as a synonym for satisfy is
> at least 450 years old. It is also spelt "satisfise."
> The term is useful because information has costs. Most attempts to find
> "optimal" solutions are counterproductive; the costs of obtaining and
> interpreting the information required almost always swampt any gains.
Thank you very much, Martin. Much better than I'd hoped for, particularly
with so few Google hits. Since google found but 9 references out of 3
billion indexed web documents, you must be a man in several hundred
million.
One of the costs involved is time- by the time the optimal solution is
found, it may no longer be optimal or the window of opportunity that
motivated the search may have closed.
Ben Zimmer
Spehro Pefhany wrote:
>
> Thank you very much, Martin. Much better than I'd hoped for, particularly
> with so few Google hits. Since google found but 9 references out of 3
> billion indexed web documents, you must be a man in several hundred
> million.
Your Googling powers are less than satisficing. I get nearly 2,000
hits.
--Ben
Spehro Pefhany
> Your Googling powers are less than satisficing. I get nearly 2,000
> hits.
Hmmm.. wonder how that happened. Must be a glitch at google.com or
something.
Bertel Lund Hansen
> 1561 J. DAUS tr. Bullinger on Apoc. (1573) 168b, That their founders
> were nourished by suckyng of a wolfe: so haue all that people wolues
> mindes, neuer satisfised with bloud, euer greedy of dominion and
> hungryng after riches.
If Iread this old text correctly, "satisfised" just means
"satisfied" here.
> 1956 H. SIMON in Psychol. Rev. LXIII. 129/2 Evidently, organisms
> adapt well enough to `satisfice'; they do not, in general, `optimize'.
The parallel between "satisfice" and "optimize" made me realize
the logic of the word. In fact it would have been clearer (to me)
from the start if it had been spelled "satisfize". The other
spelling made me think of "sacrifice" which doesn't make as much
sense.
--
Bertel, Denmark
Thomas A Lawson
<sp...@interlog.com> wrote in
message<ar348.11365$eL....@news1.bloor.is> :
>I'm reviewing a technical book proposal for a publisher and the author has
>used the word "satisfice". That word is not in AH4 or M-W, and Google
>yields a paltry 9 hits. It seems to mean something like "first reasonable
>choice", as opposed to, say, "optimal choice".
>
>Anyone else run into this word? It's not without usefulness.
>
>Best regards,
In another life in which I was a management consultant we talked about
aspects of employee satisfaction.
A "satisficing" factor was one which needed to be brought up to a
certain level. However, increasing it above that level had little
impact on overall satisfaction with remuneration/environment etc.
In short, when something "satisfices" it is "good enough". No point in
making it any better.
Unfortunately, however clear the concept, in practice it can be bloody
hard to determined the "satisficing" level.
Regards,
Tom Lawson
"There is always an easy solution to every human problem --
neat, plausible, and wrong."
- Henry Louis Mencken
Laura F Spira
combination
of "satisfy" and "suffice".
--
Laura
(emulate St. George for email)
Stephen Toogood
<sp...@interlog.com> writes
>The renowned David Squire <David....@csse.monash.edu.au> wrote:
>
>> Could I see it in a sentence?
>
>Certainly: It's necessary to satisfice on the choice of an apartment.
>
>Note: There seem to be two basic definitions (even among the few
>references). One is "first reasonable choice" and the other is "a choice
>that is (probably) less than perfect, chosen for the moment".
>
sentence as a whole?
What for example do we think about the phrase <it's necessary to
satisfice>? Is there an element of tautology here?
Then there's the preposition <on> where I would have put <with respect
to>.
We seem to agree that the use of <satisfice> is more or less confined to
technical writing in the fields of certain kinds of logic and on
organisational theory. If the word is or has become jargon, then one
tends to expect a correspondingly precise use of words in the rest of a
sentence. And if he uses a jargon word for something as un-technical as
choosing an apartment, what impenetrability can we expect when he gets
on to technical stuff?
My gut feeling is that this fellow's writing is going to be turgid and
jargon-ridden, and that the publisher ought to allow for quite a bit of
extra editorial time.
--
Stephen Toogood
R H Draney
<David....@csse.monash.edu.au> wrote:
>Spehro Pefhany wrote:
>
>> (I've been trying to come up with better ways of getting the concept
>> across of completing "most" of what is required on-time or early.
>
>Winging it? :)
The "Pareto principle" or "80-20 rule"...originally an observation
that 20% of the people in a society have 80% of the wealth, it's been
co-opted by the data processing people to mean that 20% of the total
effort expended on a task will solve 80% of the problem....r
Chris Malcolm
>The renowned Ben Zimmer <bgzi...@midway.uchicago.edu> wrote:
>> Your Googling powers are less than satisficing. I get nearly 2,000
>> hits.
>Hmmm.. wonder how that happened. Must be a glitch at google.com or
>something.
It could be that Google was satisficing.
--
Chris Malcolm c...@dai.ed.ac.uk +44 (0)131 650 3085
School of Artificial Intelligence, Division of Informatics
Edinburgh University, 5 Forrest Hill, Edinburgh, EH1 2QL, UK
[http://www.dai.ed.ac.uk/daidb/people/homes/cam/ ] DoD #205
Chris Malcolm
>We seem to agree that the use of <satisfice> is more or less confined to
>technical writing in the fields of certain kinds of logic and on
>organisational theory.
It also occurs in biology, e.g. in discussing evolution, and
philosophy, e.g. in discussing epistemics.
>If the word is or has become jargon, then one
>tends to expect a correspondingly precise use of words in the rest of a
>sentence. And if he uses a jargon word for something as un-technical as
>choosing an apartment, what impenetrability can we expect when he gets
>on to technical stuff?
Nothing jargon about it. It means no more and no less than "good
enough", which is a very straightforward, obvious, and logical
derivation from "satisfy", "satisfactory", etc.. It shouldn't need
explanation, and is a very useful word, although not often required.
>My gut feeling is that this fellow's writing is going to be turgid and
>jargon-ridden, and that the publisher ought to allow for quite a bit of
>extra editorial time.
Using my brain rather than my intestines to assess the matter, I would
expect a user of "satisfice" to be a clear and logical writer, as Herb
Simon (who was fond of the term) was.
Geoff Butler
>
>The "Pareto principle" or "80-20 rule"...originally an observation
>that 20% of the people in a society have 80% of the wealth, it's been
>co-opted by the data processing people to mean that 20% of the total
>effort expended on a task will solve 80% of the problem....r
20% of the effort solves 80% of the problem and takes 80% of
the available time. The remaining 80% of the effort solves the
remaining 20% of the problem and takes the remaining 80% of
the time.
-ler
Thomas A Lawson
<ge...@gbutler.demon.co.uk> wrote in
message<fxEK4PCP...@gbutler.demon.co.uk> :
Well, I always thought that the last 10% took 50% of the time.
But we are not too far apart. Most business would be happy with an IT
project that was only 40% to 60% over budget.
Dpm't forget: the last 5% of the problems never get solved.
j.g.ollason
: Spehro Pefhany wrote:
I think it is a coinage by Herbert Simon. It pops up from time to time in
the jargon of behavioural ecology. Some people believe that animals forage
optimally (maximize some kind of return) others believe animals satisfice
(do well enough to get by). As it happens both are wrong, but that's another
story.
John O.
Harvey V
wrote:
> I'm reviewing a technical book proposal for a publisher and the
> author has used the word "satisfice". That word is not in AH4 or
> M-W, and Google yields a paltry 9 hits. It seems to mean something
> like "first reasonable choice", as opposed to, say, "optimal
> choice".
>
> Anyone else run into this word? It's not without usefulness.
>
It's certainly been around for a few years -- I came across it in the
mid 1970s when I was studying town planning in Toronto.
At the time it was one of a number of buzz-words in sociology/social
planning circles, and we poked fun at people who used it with a
straight face.
(I wrote a gibberish patter song about planning jargon for the end-of-
year party which included the line "advocating pluralistic normative
solutions to the satsficing problem of societal objectives"....)
--
Cheers,
Harvey
John Lawler
>I'm reviewing a technical book proposal for a publisher and the author has
>used the word "satisfice". That word is not in AH4 or M-W, and Google
>yields a paltry 9 hits. It seems to mean something like "first reasonable
>choice", as opposed to, say, "optimal choice".
>Anyone else run into this word? It's not without usefulness.
I, too, ran into it in the 1970s in an academic environment (the field was
health care administration, IIRC). I'd never heard it before, so I asked
and was told that it was new jargon. I haven't really heard it much
since. It may be making the rounds of various Social Science disciplines,
as so many concepts with associated buzzwords do.
Linguistically, it's a portmanteau word, like "chortle", coined by Lewis
Carroll in 'Jabberwocky' from parts of "snort" and "chuckle" and meaning a
sound combining both. "Satisfice" combines "suffice" and "satisfy",
meaning, roughly (I'm sure there's an appropriately dense statistical
definition somewhere), satisfying some criterion/a sufficiently, though
not achieving a maximal or optimal solution.
Executive summary: "close enough for social science".
-John Lawler http://www.umich.edu/~jlawler Michigan Linguistics Dept
-----------------------------------------------------------------------
"Words are, of course, the most powerful drug used by mankind." -- Kipling
Mike Page
Hm,
NSOED has
satisfice /"satIsfVIs/ v. Also <unknown>-fise.M16. [Alt. of
SATISFY infl. by L satisfacere.]1 v.t. Satisfy. obs. exc. north.
M16. 2 v.i. Decide on and pursue a course of action that will
satisfy the minimum requirements necessary to achieve a
particular goal. M20.2 British Journal of Sociology
Prescriptions..chosen to maximise (or satisfice) on both
imperatives.
As the above shows, NSOED dates it to mid sixteenth century.
Herbert Simon I think it was who resurrected it in the 1950s to
mean choosing a course of action which satisfied the minimum
conditions in all dimensions but was not necessarily optimal.
People usually satisfice because they estimate loss of welfare
from satisficing is less than the cost of finding the optimal
solution; in real life search costs and information costs
matter.
Mike Page, BF(UU)
Let the ape escape for e-mail
Evan Kirshenbaum
> The "Pareto principle" or "80-20 rule"...originally an observation
> that 20% of the people in a society have 80% of the wealth, it's
> been co-opted by the data processing people to mean that 20% of the
> total effort expended on a task will solve 80% of the problem....r
More commonly, that 80% of the time is spent in 20% of the code, so
it's important to identify the critical regions so as to get the most
bang for your hand-optimization buck. It may not be as easy to get an
improvement there as elsewhere, but it will have more impact. This is
typically held to apply recursively, with 64% of the time spent in 4%
of the code, and so on.
--
Evan Kirshenbaum +------------------------------------
HP Laboratories |All tax revenue is the result of
1501 Page Mill Road, Building 1U |holding a gun to somebody's head.
Palo Alto, CA 94304 |Not paying taxes is against the law.
|If you don't pay your taxes, you'll
kirsh...@hpl.hp.com |be fined. If you don't pay the fine,
(650)857-7572 |you'll be jailed. If you try to
|escape from jail, you'll be shot.
http://www.kirshenbaum.net/ | P.J. O'Rourke
R H Draney
<kirsh...@hpl.hp.com> wrote:
>dado...@earthlink.net (R H Draney) writes:
>
>> The "Pareto principle" or "80-20 rule"...originally an observation
>> that 20% of the people in a society have 80% of the wealth, it's
>> been co-opted by the data processing people to mean that 20% of the
>> total effort expended on a task will solve 80% of the problem....r
>
>More commonly, that 80% of the time is spent in 20% of the code, so
>it's important to identify the critical regions so as to get the most
>bang for your hand-optimization buck. It may not be as easy to get an
>improvement there as elsewhere, but it will have more impact. This is
>typically held to apply recursively, with 64% of the time spent in 4%
>of the code, and so on.
Also that you spend 80% of your time dealing with 20% of your
customers, or that 80% of your revenue comes from 20% of your clients,
or that you can find 80% of your errors in 20% of the time it takes to
find them all, or various other manifestations...they're all
recursive, but I don't think AUE needs a discussion of Zipf's law....r