From: Philip Haddad
Subject: What do Users really want?
Date: 
Message-ID: <ba57c4f9.0411241821.d3a5e64@posting.google.com>
I have been reading PG's "Hacker's and Painters" this past week, and
he advocates the idea of startups strongly. He also makes the strong
point that to create wealth, you have to make something people want. I
have been thinking about this, and I wonder what users really want
future technology to be like. What do the user consider "cool" ideas.
I have a feeling it won't be a low-level floating-point math lib or
anything :-)
What do other programmers want? I read in "Queue" the other day that
VMs are making a comeback. Does this mean that the Java trend will
continue to expand? Wouldn't it be fairly trival to implement a VM in
CL? Since it supports Garbage Collection and dynamic typing I would
have to think that a VM in Lisp would be even more effective than one
written in Java or C. Here I am not talking about a language VM, but a
hardware-level VM.
Just some thoughts that I had bouncing around in my head. Where is the
future of user applications going. Will they be server-client models
like Graham thinks they will? Will the desktop environment slowly
dissappear over the next 50 years or so?
Overall "Hackers and Painters" is a very thought provocing book, and I
have immensely enjoyed reading every page of it.

-- 
Certum quod factum.
Philip Haddad

From: ConsideredOpinion
Subject: Re: What do Users really want?
Date: 
Message-ID: <1101359083.993890.126320@z14g2000cwz.googlegroups.com>
Which users?

If you look at the general American audience, there is more money spent
on entertainment than just about anything else.

Do something entertaining.


Philip Haddad wrote:
> I have been reading PG's "Hacker's and Painters" this past week, and
> he advocates the idea of startups strongly. He also makes the strong
> point that to create wealth, you have to make something people want.
I
> have been thinking about this, and I wonder what users really want
> future technology to be like. What do the user consider "cool" ideas.
> I have a feeling it won't be a low-level floating-point math lib or
> anything :-)
> What do other programmers want? I read in "Queue" the other day that
> VMs are making a comeback. Does this mean that the Java trend will
> continue to expand? Wouldn't it be fairly trival to implement a VM in
> CL? Since it supports Garbage Collection and dynamic typing I would
> have to think that a VM in Lisp would be even more effective than one
> written in Java or C. Here I am not talking about a language VM, but
a
> hardware-level VM.
> Just some thoughts that I had bouncing around in my head. Where is
the
> future of user applications going. Will they be server-client models
> like Graham thinks they will? Will the desktop environment slowly
> dissappear over the next 50 years or so?
> Overall "Hackers and Painters" is a very thought provocing book, and
I
> have immensely enjoyed reading every page of it.
> 
> -- 
> Certum quod factum.
> Philip Haddad
From: Stefan Scholl
Subject: Re: What do Users really want?
Date: 
Message-ID: <jxtl8ybfryxl.dlg@parsec.no-spoon.de>
On 2004-11-25 06:04:44, ConsideredOpinion wrote:

> If you look at the general American audience, there is more money spent
> on entertainment than just about anything else.
> 
> Do something entertaining.

Like voting terminals. That's always fun. :-)
From: Pascal Bourguignon
Subject: Re: What do Users really want?
Date: 
Message-ID: <87k6sak4qy.fsf@thalassa.informatimago.com>
·············@gmail.com (Philip Haddad) writes:

> I have been reading PG's "Hacker's and Painters" this past week, and
> he advocates the idea of startups strongly. He also makes the strong
> point that to create wealth, you have to make something people want. I
> have been thinking about this, and I wonder what users really want

That's easy: sex, beer, sex, food, sex...


> future technology to be like. What do the user consider "cool" ideas.

Anything that provides the above mentionned.


> I have a feeling it won't be a low-level floating-point math lib or
> anything :-)
> What do other programmers want? 

Yes, at first they may seem kind of aliens, in wanting powerful
computers, and lisp system, to be able to write nice programs.  But
actually, they're just like the other users, seeking the above
mentionned, only more pervertly seeking it via detourned means: making
a lot of money selling their programs, then use that money to get the
above mentionned.

 
> [...]
> Just some thoughts that I had bouncing around in my head. Where is the
> future of user applications going. Will they be server-client models
> like Graham thinks they will? Will the desktop environment slowly
> dissappear over the next 50 years or so?

Well, since keeping in front of the screen of a computer at a desktop,
for users (but day-traders, they're as pervert as the programmers), is
not exactly the best way to get the above mentionned, I'd say that
yes, it will disappear.


I'd bet on embedded AI "robotics", applied to this kind of hardware:
    http://www.realdoll.com/

A first step is: 
    http://www.3g.co.uk/PR/August2004/8226.htm

(Also, have a look at the "Chobits" Japanese anime).


> Overall "Hackers and Painters" is a very thought provocing book, and I
> have immensely enjoyed reading every page of it.

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
The world will now reboot; don't bother saving your artefacts.
From: Geoffrey Summerhayes
Subject: Re: What do Users really want?
Date: 
Message-ID: <8Vdpd.39334$Ro.1716363@news20.bellglobal.com>
"Pascal Bourguignon" <····@mouse-potato.com> wrote in message ···················@thalassa.informatimago.com...
> ·············@gmail.com (Philip Haddad) writes:
>
>> I have been reading PG's "Hacker's and Painters" this past week, and
>> he advocates the idea of startups strongly. He also makes the strong
>> point that to create wealth, you have to make something people want. I
>> have been thinking about this, and I wonder what users really want
>
> That's easy: sex, beer, sex, food, sex...

A computer programmer happens across a frog in the road. The frog
pipes up, "I'm really a beautiful princess and if you kiss me, I'll
stay with you for a week". The programmer shrugs his shoulders and puts
the frog in his pocket.
A few minutes later, the frog says "OK, OK, if you kiss me, I'll give
you great sex for a week". The programmer nods and puts the frog back
in his pocket.

A few minutes later, "Turn me back into a princess and I'll give you
great sex for a whole year!". The programmer smiles and walks on.

Finally, the frog says, "What's wrong with you? I've promised you
great sex for a year from a beautiful princess and you won't even kiss
a frog?"

"I'm a programmer," he replies. "I don't have time for
sex.... But a talking frog is pretty neat."

--

Geoff
From: Mark McConnell
Subject: Re: What do Users really want?
Date: 
Message-ID: <d3aed052.0411290943.42989856@posting.google.com>
"Geoffrey Summerhayes" <·······@NhOoStPmAaMil.com> wrote in message news:<······················@news20.bellglobal.com>...
> "Pascal Bourguignon" <····@mouse-potato.com> wrote in message ···················@thalassa.informatimago.com...
> > ·············@gmail.com (Philip Haddad) writes:
> >
> >> I have been reading PG's "Hacker's and Painters" this past week, and
> >> he advocates the idea of startups strongly. He also makes the strong
> >> point that to create wealth, you have to make something people want. I
> >> have been thinking about this, and I wonder what users really want
> >
> > That's easy: sex, beer, sex, food, sex...
> 
> A computer programmer happens across a frog in the road. The frog
> pipes up, "I'm really a beautiful princess and if you kiss me, I'll
> stay with you for a week". The programmer shrugs his shoulders and puts
> the frog in his pocket.
> A few minutes later, the frog says "OK, OK, if you kiss me, I'll give
> you great sex for a week". The programmer nods and puts the frog back
> in his pocket.
> 
> A few minutes later, "Turn me back into a princess and I'll give you
> great sex for a whole year!". The programmer smiles and walks on.
> 
> Finally, the frog says, "What's wrong with you? I've promised you
> great sex for a year from a beautiful princess and you won't even kiss
> a frog?"
> 
> "I'm a programmer," he replies. "I don't have time for
> sex.... But a talking frog is pretty neat."

From an old Sylvia cartoon:

A: This article says they surveyed 100,000 men to find out what they
really like in bed.

B: So, what do men really like in bed?

A: Breakfast.

B: (I knew that, deep down.)
From: Steven M. Haflich
Subject: Re: What do Users really want?
Date: 
Message-ID: <uNhpd.26781$6q2.23361@newssvr14.news.prodigy.com>
Pascal Bourguignon wrote:

> That's easy: sex, beer, sex, food, sex...

Money.  With money you can buy all the others.
From: lin8080
Subject: Re: What do Users really want?
Date: 
Message-ID: <41A894E4.C01B58B0@freenet.de>
Pascal Bourguignon schrieb:
> 
> ·············@gmail.com (Philip Haddad) writes:
> 
> > I have been reading PG's "Hacker's and Painters" this past week, and
> > he advocates the idea of startups strongly. He also makes the strong
> > point that to create wealth, you have to make something people want. I
> > have been thinking about this, and I wonder what users really want

> That's easy: sex, beer, sex, food, sex...

You seem to forget one important point: money. But the rest: right.
Always more and more...

So the question is better formed: what will be payed tomorrow?
On trend is in sight, the magic word is: smaller *.*
And in this way CL is 2MB, very small.

What is now can be named multimedia, full communicatin, security. 
Looking at Bills plans the privat money-giver seems the goal. 

But realistic, the next step can be called: speak to me. An importand
point not only for nowadays desktop-users. This may be able to shut down
the classical desktop and gives "chips" a never-before-seen mobility in
our world, say everywhere. Industry should be prepared to make sensors.

The next is the problem about technics. The wishes may be 3D,
holographics, nanos and all that stuff one can see in a scifi movie.
All this makes one thing: huge data-streams, endless search-spaces,
abstractions everywhere (assuming, speed goes on).

To name an example: assume you have a 300kb lisp-package of a
software-tool and you will be able to bring the runtime-version down to
80kb without changing the functionality. This is the breakpoint, I mean. 
Well, call it VM, but better make the syntax for this. :)

And still there is that thing called: do what I mean. 

Strategic viewpoint to lisp: you have the socalled primitives (like car,
cons) and the same time you have over the time addons to do the
high-level-things. I would start in this: reform the syntax of lisp,
main for portability and interface-managing, or put it away from things
called temporarly hack. Even these days I read in CL src init.lisp
(leads to such things):

;; Caution! For bootstrapping purposes (recognizable with #-COMPILER)
some
;; of the functions have to be written in a more primitive Lisp
;; (without do, do*, case).

(point out to think about)

lin

my babylon translater diappeared :(
From: Tim Bradshaw
Subject: Re: What do Users really want?
Date: 
Message-ID: <1101396779.936348.272500@f14g2000cwb.googlegroups.com>
Philip Haddad wrote:
> I have been reading PG's "Hacker's and Painters" this past week, and
> he advocates the idea of startups strongly. He also makes the strong
> point that to create wealth, you have to make something people want.
I
> have been thinking about this, and I wonder what users really want
> future technology to be like. What do the user consider "cool" ideas.
> I have a feeling it won't be a low-level floating-point math lib or
> anything :-)

1. Things that work reliably and simply.
2. Money.
3. Other things.

If you can do (1) you will get lots of (2) and then you can get
(almost all of) the things in (3).  We (computing professionals) are
dismally and absolutely failing to do (1).  In fact we're gettign
further away from (1) very rapidly indeed.

--tim
From: Fred Gilham
Subject: Re: What do Users really want?
Date: 
Message-ID: <u7vfbtub5x.fsf@snapdragon.csl.sri.com>
Tim Bradshaw wrote:
> 1. Things that work reliably and simply.
> 2. Money.
> 3. Other things.
>
> If you can do (1) you will get lots of (2) and then you can get
> (almost all of) the things in (3).  We (computing professionals) are
> dismally and absolutely failing to do (1).  In fact we're gettign
> further away from (1) very rapidly indeed.

The premise I've gone by for as long as I've been involved with
computers is that 1) should be accomplished by automating the
programming process.  Thus using a compiler is better in terms of 1)
than writing assembly language; Lisp is better than C because of
garbage collection; and so on.  The less that stands between the
programmer and the realization of the programmer's idea, the better it
will be in terms of 1), and my assumption is that automating the
programming process as much as possible will facilitate this.

Is this a reasonable assumption?

(I still remember sitting in a design meeting talking about a system
we were building.  The question of design criteria came up.  I said,
"Simplicity."  There was silence for a little while, then the
discussion moved on.  Simplicity was never heard from again.  Needless
to say the design was not simple.)

-- 
Fred Gilham ······@csl.sri.com || "If I thought there was anything at
all in your arguments, I should have to be not only a theist, but an
Episcopalian to boot," he said, after one interchange, reckoning that
since Episcopalianism was, in his book, that than which nothing could
be worse, this was an effective reductio ad absurdum. - J. R. Lucas
From: Svein Ove Aas
Subject: Re: What do Users really want?
Date: 
Message-ID: <co535b$q4d$1@services.kq.no>
Fred Gilham wrote:

> 
> Tim Bradshaw wrote:
>> 1. Things that work reliably and simply.
>> 2. Money.
>> 3. Other things.
>>
>> If you can do (1) you will get lots of (2) and then you can get
>> (almost all of) the things in (3).  We (computing professionals) are
>> dismally and absolutely failing to do (1).  In fact we're gettign
>> further away from (1) very rapidly indeed.
> 
> The premise I've gone by for as long as I've been involved with
> computers is that 1) should be accomplished by automating the
> programming process.  Thus using a compiler is better in terms of 1)
> than writing assembly language; Lisp is better than C because of
> garbage collection; and so on.  The less that stands between the
> programmer and the realization of the programmer's idea, the better it
> will be in terms of 1), and my assumption is that automating the
> programming process as much as possible will facilitate this.
> 
> Is this a reasonable assumption?
> 
Empirical evidence appears to support your conclusion.

> (I still remember sitting in a design meeting talking about a system
> we were building.  The question of design criteria came up.  I said,
> "Simplicity."  There was silence for a little while, then the
> discussion moved on.  Simplicity was never heard from again.  Needless
> to say the design was not simple.)
> 
That's telling. Some people are aware of the importance of simplicity, but
they usually don't make the decisions. Java is, at least theoretically, a
simpler environment than C++; that said, Java programs/frameworks tend to
be far more complex than C++ ones. Why is that?

- Complexity equals job safety?
- Complexity looks good?
- Bad habits?
- Complexity *feels* good?
- Premature generalization?


For another anecdote: A year ago I was supposed to write a simple database
program, in Java. So I did, but I wound up writing a (much more complex)
pseudo-XML backend that was hard to debug and didn't actually buy me
anything!

I've learnt better; many of my brethren haven't, and appear unable to
understand why they shouldn't do that. Of course, I probably wouldn't have
listened either; I'm not quite sure what made me change my mind. c2.com
definitely helped.
From: Matthias
Subject: Re: What do Users really want?
Date: 
Message-ID: <36wbrdlslfq.fsf@hundertwasser.ti.uni-mannheim.de>
Svein Ove Aas <·········@aas.no> writes:
> That's telling. Some people are aware of the importance of simplicity, but
> they usually don't make the decisions. Java is, at least theoretically, a
> simpler environment than C++; that said, Java programs/frameworks tend to
> be far more complex than C++ ones. Why is that?
> 
> - Complexity equals job safety?
> - Complexity looks good?
> - Bad habits?
> - Complexity *feels* good?
> - Premature generalization?

- You have to understand a problem really well until you see the
  simple solution.
From: mark
Subject: Re: What do Users really want?
Date: 
Message-ID: <eEzpd.152992$R05.101454@attbi_s53>
Matthias wrote:
> Svein Ove Aas <·········@aas.no> writes:
> 
>>That's telling. Some people are aware of the importance of simplicity, but
>>they usually don't make the decisions. Java is, at least theoretically, a
>>simpler environment than C++; that said, Java programs/frameworks tend to
>>be far more complex than C++ ones. Why is that?
>>
>>- Complexity equals job safety?
>>- Complexity looks good?
>>- Bad habits?
>>- Complexity *feels* good?
>>- Premature generalization?
> 
> 
> - You have to understand a problem really well until you see the
>   simple solution.
> 
what would you spend money on and how much? (for software)
From: Julian Stecklina
Subject: Re: What do Users really want?
Date: 
Message-ID: <86wtw81rl0.fsf@goldenaxe.localnet>
mark <·············@_myway.com> writes:

> what would you spend money on and how much? (for software)

30 EUR for threads in SBCL on FreeBSD.
20 EUR for alien callbacks in SBCL on FreeBSD. 

(I cannot think of any other software at the moment. I am still
horrified what I would have to pay to get an ACL+CLIM license...)

Regards,
-- 
                    ____________________________
 Julian Stecklina  /  _________________________/
  ________________/  /
  \_________________/  LISP - truly beautiful
From: Chris Capel
Subject: Re: What do Users really want?
Date: 
Message-ID: <10qftm4fjj9bna2@corp.supernews.com>
Svein Ove Aas wrote:

> That's telling. Some people are aware of the importance of simplicity, but
> they usually don't make the decisions. Java is, at least theoretically, a
> simpler environment than C++; that said, Java programs/frameworks tend to
> be far more complex than C++ ones. Why is that?
> 
> - Complexity equals job safety?
> - Complexity looks good?
> - Bad habits?
> - Complexity *feels* good?
> - Premature generalization?
> 
> 
> For another anecdote: A year ago I was supposed to write a simple database
> program, in Java. So I did, but I wound up writing a (much more complex)
> pseudo-XML backend that was hard to debug and didn't actually buy me
> anything!
> 
> I've learnt better; many of my brethren haven't, and appear unable to
> understand why they shouldn't do that. Of course, I probably wouldn't have
> listened either; I'm not quite sure what made me change my mind. c2.com
> definitely helped.

If I may prattle on a bit, I think mastering conceptual generalization is
the first step towards coding nirvana. It's the first aesthetic lesson of
the junior coder. Once that is mastered, one might be said to be an
intermediate student. The next step is syntactic generalization and
manipulation. This, once begun, greatly enhances the rate of the student's
progress in seeing new conceptual generalizations. For instance, how would
one go about programming a dataflow system (like Cells) in a language like
Java? It would be fairly difficult, unobvious, and verbose, at best.
Perhaps someone might see the advantage of it (especially if they were
inspired by more dynamic languages), but only with knowledge of language
with syntactic flexibility can one really appreciate it.

So what's the third lesson? I dunno. I probably have the progression
organized all wrong.

Chris Capel
From: Tim Bradshaw
Subject: Re: What do Users really want?
Date: 
Message-ID: <1101468873.058012.221400@z14g2000cwz.googlegroups.com>
Fred Gilham wrote:
> The premise I've gone by for as long as I've been involved with
> computers is that 1) should be accomplished by automating the
> programming process.  Thus using a compiler is better in terms of 1)
> than writing assembly language; Lisp is better than C because of
> garbage collection; and so on.  The less that stands between the
> programmer and the realization of the programmer's idea, the better
it
> will be in terms of 1), and my assumption is that automating the
> programming process as much as possible will facilitate this.
>
> Is this a reasonable assumption?
>

I think so.  However, as I've said before, I think (now) that issues
such as programming language choice turn out to be fairly low on the
list of problems.  It makes a difference of course (and I think the
single largest difference is GC) but not as much as programmers
generally believe.

--tim
From: Pascal Costanza
Subject: Re: What do Users really want?
Date: 
Message-ID: <co7pua$3fc$1@newsreader2.netcologne.de>
Tim Bradshaw wrote:

> I think (now) that issues
> such as programming language choice turn out to be fairly low on the
> list of problems.  It makes a difference of course (and I think the
> single largest difference is GC) but not as much as programmers
> generally believe.

I think another property of a programming language that's still 
important is how easy it is to change important design decisions very 
late in the game. Many of the programming concepts around work well only 
"most" of the time. An important question is how easily one can get 
around unforeseen barriers without having to start from scratch when a 
concrete approach breaks.

Of course, such a property is much harder to pinpoint than garbage 
collection.

It also seems to me that at least either closures or macros should be 
available in a language.


Pascal

-- 
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
From: Chris Capel
Subject: Re: What do Users really want?
Date: 
Message-ID: <10qft4ea8u0te38@corp.supernews.com>
Tim Bradshaw wrote:

> I think (now) that issues
> such as programming language choice turn out to be fairly low on the
> list of problems.  It makes a difference of course (and I think the
> single largest difference is GC) but not as much as programmers
> generally believe.

Well, it depends on which programmers you're looking at. A few people manage
to be successful in small companies that have lots and lots of raw ability
and sophistication. Many of the problems that I think you're alluding to
are a result of so much software being produced by big companies with poor
middle-management (and often poor upper management). Short-sightedness,
groupthink, problems of scale (the Mythical Man Month). Though I think that
interoperability nightmares play a big part in the poor state of the art,
and those are simply inherent in the field. Aren't they? It's the tension
between standards and progress.

But one thing I don't really have a sense of is which mode dominates. Are
there more small independents out there doing good work, or does more
software come out of the big shops? And to what degree do the small shops
have terrible practices, and how many big ones know what they're doing?

And what about the future? What kind of organization should we be trying for
to help the industry achieve the kind of things that it's failed at so far?

I really don't think any language can even begin to solve these problems.
Until education becomes better (either formal or informal), the crappiest
code will be written in the most popular language.

Hmm.

Chris Capel
From: lin8080
Subject: Re: What do Users really want?
Date: 
Message-ID: <41AAEEC1.56AFF770@freenet.de>
> Tim Bradshaw wrote:

> I really don't think any language can even begin to solve these problems.
> Until education becomes better (either formal or informal), the crappiest
> code will be written in the most popular language.

> Hmm.

Why wait for others? 
It is up to you, to enter new areas. 
You see the aim, you can go the way, you should make a step towards.

Make that step. 

lin

(did t.linus guess what he did to the world?)
From: Jacek Generowicz
Subject: Re: What do Users really want?
Date: 
Message-ID: <tyfr7mgevhz.fsf@pcepsft001.cern.ch>
Fred Gilham <······@snapdragon.csl.sri.com> writes:

> (I still remember sitting in a design meeting talking about a system
> we were building.  The question of design criteria came up.  I said,
> "Simplicity."  There was silence for a little while, then the
> discussion moved on.  Simplicity was never heard from again.  Needless
> to say the design was not simple.)

But simplicity can mean wildly different things to different
people.

I consider finding a good abstraction and letting the abstraction
generate a whole range of different realizations of that abstraction,
to lead to simple design. I consider looking for "symmerties" in my
programs and eliminating special cases by making them fit into some
symmetric pattern, into some parametrization of an abstraction, to be
simple design.

I have endless arugments with a colleague on this matter. He considers
it far simpler to code up each variation separately as a special case.

I consider code generation to lead to simple design; he considers
wirting the whole lot out by hand to be simpler. I insist or storing
only the gode generator in CVS; he wants me to store the generated
code in CVS because including the code generation step in the build
process is complicated while compiling what you find in CVS is simple.


I guess it's a long-term-perspective vs short-term-perspective thing.
From: Adrian Kubala
Subject: Re: What do Users really want?
Date: 
Message-ID: <slrncqeniu.ccn.adrian@sixfingeredman.net>
Jacek Generowicz <················@cern.ch> schrieb:
> Fred Gilham <······@snapdragon.csl.sri.com> writes:
>> (I still remember sitting in a design meeting talking about a system
>> we were building.  The question of design criteria came up.  I said,
>> "Simplicity."  There was silence for a little while, then the
>> discussion moved on.  Simplicity was never heard from again.
>> Needless to say the design was not simple.)
>
> But simplicity can mean wildly different things to different people.

"Once and only once" seems to me like a good measure of simplicity that
matches most people's intuition. At least, I think complexity
necessarily corresponds to the opposite of OAOO because complex = more
interconnections = more information spread out and duplicated in the
interfaces between parts. Sure, this just turns judging simplicity into
the equally hard job of weighing the costs of different kinds of
duplication, but it still provides a more objective language for
discussion.
From: William Bland
Subject: Re: What do Users really want?
Date: 
Message-ID: <pan.2004.11.26.17.30.11.862736@abstractnonsense.com>
On Fri, 26 Nov 2004 11:48:40 +0100, Jacek Generowicz wrote:

> But simplicity can mean wildly different things to different
> people.
> 
...
> 
> I guess it's a long-term-perspective vs short-term-perspective thing.

You are correct.  Your colleague is wrong.  I have similar arguments with
a colleague at my place of work.  You have my sympathies.

Cheers,
	Bill.
From: Jacek Generowicz
Subject: Re: What do Users really want?
Date: 
Message-ID: <tyfllcldp8a.fsf@pcepsft001.cern.ch>
William Bland <·······@abstractnonsense.com> writes:

> On Fri, 26 Nov 2004 11:48:40 +0100, Jacek Generowicz wrote:
> 
> > But simplicity can mean wildly different things to different
> > people.
> > 
> ...
> > 
> > I guess it's a long-term-perspective vs short-term-perspective thing.
> 
> You are correct.  Your colleague is wrong.

Unfortunately my colleague is more accurately described as my
boss. Therefore he is right and I am wrong, by definition. :-(
From: Peter Seibel
Subject: Re: What do Users really want?
Date: 
Message-ID: <m3is7oo811.fsf@javamonkey.com>
Jacek Generowicz <················@cern.ch> writes:

> William Bland <·······@abstractnonsense.com> writes:
>
>> On Fri, 26 Nov 2004 11:48:40 +0100, Jacek Generowicz wrote:
>> 
>> > But simplicity can mean wildly different things to different
>> > people.
>> > 
>> ...
>> > 
>> > I guess it's a long-term-perspective vs short-term-perspective thing.
>> 
>> You are correct.  Your colleague is wrong.
>
> Unfortunately my colleague is more accurately described as my
> boss. Therefore he is right and I am wrong, by definition. :-(

Well, on the bright side, you keep posting stuff like this in public
places and he won't be your boss for long. ;-)

-Peter

-- 
Peter Seibel                                      ·····@javamonkey.com

         Lisp is the red pill. -- John Fraser, comp.lang.lisp
From: Michael Sullivan
Subject: Re: What do Users really want?
Date: 
Message-ID: <1gnzj6n.d70nts1kn7me7N%mes@panix.com>
Jacek Generowicz <················@cern.ch> wrote:

> Fred Gilham <······@snapdragon.csl.sri.com> writes:
> 
> > (I still remember sitting in a design meeting talking about a system
> > we were building.  The question of design criteria came up.  I said,
> > "Simplicity."  There was silence for a little while, then the
> > discussion moved on.  Simplicity was never heard from again.  Needless
> > to say the design was not simple.)
> 
> But simplicity can mean wildly different things to different
> people.

> I consider finding a good abstraction and letting the abstraction
> generate a whole range of different realizations of that abstraction,
> to lead to simple design. I consider looking for "symmerties" in my
> programs and eliminating special cases by making them fit into some
> symmetric pattern, into some parametrization of an abstraction, to be
> simple design.

> I have endless arugments with a colleague on this matter. He considers
> it far simpler to code up each variation separately as a special case.

What you are talking about is simplicity.  What he is talking about is
ease.  Simplicity of design does not make the design easier, it makes it
much harder, assuming you intend to get all the important functionality
in there.  But when you're done, the product will be much better.
 
Your colleague is basically advocating an option which spends no time on
design at all.  Of course this looks easier in the short run.  The
question is whether the trade-offs involved are worthwhile.  For some
quick and dirty hack, they may be.  But as we all know, many a quick and
dirty hack ends up being at the center of something big and widely
re-used.


Michael

-- 
"Every gun that is made, every warship launched, every rocket fired,
signifies in the final sense a theft from those who hunger and are not
fed, those who are cold and are not clothed.  -- Dwight Eisenhower 
         "In Christ there is no killing" -- St. Patrick
From: Svein Ove Aas
Subject: Re: What do Users really want?
Date: 
Message-ID: <co5298$ka0$1@services.kq.no>
Tim Bradshaw wrote:

> Philip Haddad wrote:
>> I have been reading PG's "Hacker's and Painters" this past week, and
>> he advocates the idea of startups strongly. He also makes the strong
>> point that to create wealth, you have to make something people want.
> I
>> have been thinking about this, and I wonder what users really want
>> future technology to be like. What do the user consider "cool" ideas.
>> I have a feeling it won't be a low-level floating-point math lib or
>> anything :-)
> 
> 1. Things that work reliably and simply.
> 2. Money.
> 3. Other things.
> 
> If you can do (1) you will get lots of (2) and then you can get
> (almost all of) the things in (3).  We (computing professionals) are
> dismally and absolutely failing to do (1).  In fact we're gettign
> further away from (1) very rapidly indeed.

Time to break the generally dismal view of this thread, I think.
_We_ are Lisp programmers; I'm still a studnent, so I don't qualify as a
computing professional - thankfully.

Lisp is *all about* writing things that work reliably and simply. If the
rest of society doesn't want to follow us, that's.. hmm, actually it's our
problem too, but at least we know there's a better way.

Therefore, we have far better reasons to despair than everyone else. Yeah,
that'll cheer them up.
From: Svein Ove Aas
Subject: Re: What do Users really want?
Date: 
Message-ID: <co52m5$ka0$3@services.kq.no>
Tim Bradshaw wrote:

> Philip Haddad wrote:
>> I have been reading PG's "Hacker's and Painters" this past week, and
>> he advocates the idea of startups strongly. He also makes the strong
>> point that to create wealth, you have to make something people want.
> I
>> have been thinking about this, and I wonder what users really want
>> future technology to be like. What do the user consider "cool" ideas.
>> I have a feeling it won't be a low-level floating-point math lib or
>> anything :-)
> 
> 1. Things that work reliably and simply.
> 2. Money.
> 3. Other things.
> 
> If you can do (1) you will get lots of (2) and then you can get
> (almost all of) the things in (3).  We (computing professionals) are
> dismally and absolutely failing to do (1).  In fact we're gettign
> further away from (1) very rapidly indeed.

Time to break the generally dismal view of this thread, I think.
_We_ are Lisp programmers; I'm still a student, so I don't qualify as a
computing professional - thankfully.

Lisp is *all about* writing things that work reliably and simply. If the
rest of society doesn't want to follow us, that's.. hmm, actually it's our
problem too, but at least we know there's a better way.

Therefore, we have far better reasons to despair than everyone else. Yeah,
that'll cheer them up.
From: Wade Humeniuk
Subject: Re: What do Users really want?
Date: 
Message-ID: <9Impd.206953$9b.78664@edtnps84>
Philip Haddad wrote:

> Just some thoughts that I had bouncing around in my head. Where is the
> future of user applications going. Will they be server-client models
> like Graham thinks they will? Will the desktop environment slowly
> dissappear over the next 50 years or so?

 From the business standpoint, the desktop's primary function is to
collect and organize information.  Data entry, document prep,
spreadsheets and presentations take up a large proportion of the time
of the office worker.  We will probably see more work done by the
computer in formatting and presenting the information.  Some kind
of template based libraries (with hopefully some AI stuff) to
create reports and documentation.  Also more automated collection
of data, through computer surveillence (re: RFID tags,  electronic
transactions).  That means more mutual agreements as to what the
machinery of business is.  Also the more widespread use of digital
certificates to verify that you are you.

More industrial robots.  The creation of robots capable of making
clothing (something that is now only possible with very simple designs).
Digital paper, the merging of the paper interface with computers.
Most people are just consumers of information/entertainment so
make it easier to access informaton (in this case the breadth
of information available on the web in book form will be a killer
app). More viruses, more advertising (personalized), the elimination
of paper newspapers, the almost total automation of driving your
car.  More problems with sendentary (obesity, lack of exercise)
lifesyles caused by having to do less.

The single fixed desktop will disappear.  People will have PDA like devices,
but smaller, that when they approach a fixed station (with some kind of
display, keypad) they can utilize that station's capabilities using
the total of the informaton stored in their personal PDA (or stored
somewhere).  This PDA will work with a variety of devices, such as
an electronic movie outlet, downloading movies on demand.  It will also
keep your medical data, available for any doctor or hospital to use
(at your request of course).  It might even be able to interact with
things like MRI machines, keeping the scan with all its pertinent
info.  Thus people will become a distributed form of database/intelligent
agents, merging with their PDAs in society's official concept of a
person.

Wade

> Overall "Hackers and Painters" is a very thought provocing book, and I
> have immensely enjoyed reading every page of it.
> 
From: Philip Haddad
Subject: Re: What do Users really want?
Date: 
Message-ID: <ba57c4f9.0411251726.75439be5@posting.google.com>
> and I wonder what users really want

When I said "users" here I meant the people that use our apps.
Although I did ask what programmers want, I meant the emphises to be
on users themselves. Programmers I think will want what their users
want. By this I mean that programmers will have to design what the
public demands. What is that?

BTW, as a side note, I found that the article in "Queue" was 4+ months
old.

Graham also mentions that in his startup Viaweb, it was useful for
them to be using Lisp, because no other competitor knew how to program
in it. As well as being the best language around of course ;)

What are the trends that we may be seeing now that could be telling us
things for the future? I mean what are users using now that will
continue in the future?

Just thought I would clear some of those items up.

-- 
Certum quod factum.
Philip Haddad
From: Trent Buck
Subject: Re: What do Users really want?
Date: 
Message-ID: <20041126202349.187352f4@harpo.marx>
Up spake Philip Haddad:
> Graham also mentions that in his startup Viaweb, it was useful for
> them to be using Lisp, because no other competitor knew how to program
> in it. As well as being the best language around of course ;)

You should put those statements in logical order:
  Lisp was better-suited for the task.          --- (1)
  Graham new (1) and his competitors did not.   --- (2)
  Graham had a technical advantage.             --- (3)

  (1) AND (2) IMPLIES (3).

An important observation Graham makes is that he was writing a
server-side application -- deploying to a few servers under his control,
rather than many desktops.  This strengthens (1).

-- 
-trent
How, Joe wondered, can a man have courage without faith, without belief?
Burroughs believed in nothing, and yet there he sat stubborn as Luther.
From: Trent Buck
Subject: Re: What do Users really want?
Date: 
Message-ID: <20041126203553.1933218c@harpo.marx>
Up spake Trent Buck:
>   Graham new (1) and his competitors did not.   --- (2)

s/new/knew/
From: Mario S. Mommer
Subject: Re: What do Users really want?
Date: 
Message-ID: <fzhdncxyct.fsf@germany.igpm.rwth-aachen.de>
·············@gmail.com (Philip Haddad) writes:
> I mean what are users using now that will continue in the future?

Broken, bloated, and buggy software with more and more malware. They
seem to absolutely love it.
From: Michael Sullivan
Subject: Re: What do Users really want?
Date: 
Message-ID: <1gnzjny.ujn7et7znlqxN%mes@panix.com>
Mario S. Mommer <········@yahoo.com> wrote:

> ·············@gmail.com (Philip Haddad) writes:
> > I mean what are users using now that will continue in the future?
 
> Broken, bloated, and buggy software with more and more malware. They
> seem to absolutely love it.

Actually, many users hate it.  The problem is that they hate these
things even more:

1) not being able to seamlessly exchange files with everybody else
2) worrying that small company X who wrote their software will go out
   of business and leave them in the lurch.
3) paying more money without being absolutely certain the product will
   be superior, because they already know someone who has it.
4) not being able to get tech support from random "computer people"

Figure out how to deal with those 4 issues without being a huge behemoth
of a company, and you will have something interesting.


Michael

-- 
"Every gun that is made, every warship launched, every rocket fired,
signifies in the final sense a theft from those who hunger and are not
fed, those who are cold and are not clothed.  -- Dwight Eisenhower 
         "In Christ there is no killing" -- St. Patrick
From: Philip Haddad
Subject: Re: What do Users really want?
Date: 
Message-ID: <ba57c4f9.0411291315.412fc51c@posting.google.com>
> Actually, many users hate it.  The problem is that they hate these
> things even more:
> 
> 1) not being able to seamlessly exchange files with everybody else
> 2) worrying that small company X who wrote their software will go out
>    of business and leave them in the lurch.
> 3) paying more money without being absolutely certain the product will
>    be superior, because they already know someone who has it.
> 4) not being able to get tech support from random "computer people"

Amen to that! My boss was worryed about buying a computer from small
company X and not being able to get a "techie" to help her with it. I
think that users may need to learn a little more about computers these
days, it really doesn't take a geeky programmer to figure it out
anymore ;-)

> Figure out how to deal with those 4 issues without being a huge behemoth
> of a company, and you will have something interesting.

Maybe I will...........
then again maybe I won't, but I have a feeling someone will sooner or
later.
 
> 
> Michael

-- 
Certum quod factum.
Philip Haddad
From: lin8080
Subject: Re: What do Users really want?
Date: 
Message-ID: <41A89AF9.A0CCFD18@freenet.de>
Philip Haddad schrieb:

> What are the trends that we may be seeing now that could be telling us
> things for the future? I mean what are users using now that will
> continue in the future?

Hmmm...
/start headsick
Do you think a pair of trousers with a klick-buttom for temporary
transparents will make it? Better more buttons? Include a calendar? A
small Phonebook?
/end headsick

What peoples realy wants is a secret in their heads. Find and
produce....

lin
From: Vladimir Sedach
Subject: Re: What do Users really want?
Date: 
Message-ID: <87d5y26t4a.fsf@shawnews.cg.shawcable.net>
·············@gmail.com (Philip Haddad) writes:

> I have been reading PG's "Hacker's and Painters" this past week, and
> he advocates the idea of startups strongly. He also makes the strong
> point that to create wealth, you have to make something people want.

Paul Graham is a master at stating the obvious and getting attention
for it, which says a lot more about his audience than about PG
himself. I think it's been pretty clear for the past 200 years that if
you want to get rich, the surest way is to create your own company
that sells something people want. Of course, people don't know what
they want (this is the real answer to your question, btw), so you have
to convince them they want what you're selling (which is really a lot
harder than it sounds). I haven't read _Hackers and Painters_, but
what does PG say about venture capitalists? The surest way not to make
money from a start-up seems to be to give controlling interest to VCs
and then let them bring in their own "professional" management (most
of the time, they're professionals only in the sense that they're
overcompensated).

> What do other programmers want?

A low-level floating-point math lib.

>I read in "Queue" the other day that VMs are making a comeback. Does
>this mean that the Java trend will continue to expand?

Wow, those guys at the Queue are really staying on top of the news,
aren't they?

> Wouldn't it be fairly trival to implement a VM in CL? Since it
> supports Garbage Collection and dynamic typing I would have to think
> that a VM in Lisp would be even more effective than one written in
> Java or C. Here I am not talking about a language VM, but a
> hardware-level VM.

I don't think users want VMs. I don't even think most programmers want
VMs. The only people who are really pushing VMs now are Sun and
Microsoft, and the only reason they're doing it is to consolidate
control around their "software platforms" and whatnot. Notice how the
above business lesson applies: if you want to make money, you have to
have control over whatever it is you are selling, and you have to fool
everyone into thinking that's what they want.

> Where is the future of user applications going.

Mobile phones. In the future everything (that the "users" care about
anyway) is going to be on mobiles.

Vladimir
From: Trent Buck
Subject: Re: What do Users really want?
Date: 
Message-ID: <20041126063106.72772003@harpo.marx>
Quoth Vladimir Sedach on or about 2004-11-25:
> > What do other programmers want?

Dere Hogfather, for Hogswatch I would lyke...
> A low-level floating-point math lib.

> If you want to make money, you have to
> have control over whatever it is you are selling, and you have to fool
> everyone into thinking that's what they want.

Unless you're selling a service, although I suppose that constitutes
`control' over a skill set / employee.

> > Where is the future of user applications going.
> 
> Mobile phones. In the future everything (that the "users" care about
> anyway) is going to be on mobiles.

And HTML interfaces for everything :-/

-trent