Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[Vester Thacker]

On 5/12/07, lucio@proxima.alt.za <lucio@proxima.alt.za> wrote:
>
> Don't believe the rationalisations, consider instead the individual
> objectives.  Who are the real contributors and who are the limelight
> seekers?  Plan 9 seems to be resisting a swing from elitism to
> communism.

Thanks, but I'll pass on the kool aid. Why make it an us and them
scenario? But for the sake of argument I'll follow your example.

I don't care whose name is on it. I don't care who is making money
with Plan 9. More power to the folks behind Rangboom[1]. If I have an
agenda, it would be Plan 9 education and having fun! Enthusiasts are
not limelight seekers. In short, we're people who enjoy Plan 9, and
would like to help in some possible way. We're not the great unwashed
masses.

If I am skilled enough to do it, I might be able to setup a subversion
server and a web portal for regular folks that want to learn and
contribute. Sometimes it seems that it does no good to talk about it.
Just do it.  Volunteers welcomed to join. BTW, if someone does this
before I do, then more power to you; I'll follow your lead.

> Who will gain the most from the swing when the reality
> check shows that it has been from a meritocracy to mediocrity?

I hope the end users gain the most. What is the purpose of writing an
operating system if you don't expect average folks to use it or
develop for it?

Vester Thacker
Amatuers do it for love. :)

[1] http://www.rangboom.com !!! :)

P.S. - This is my last posting with my real name. I'll use an alias from now on.

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad

[lucio]

> I hope the end users gain the most. What is the purpose of writing an
> operating system if you don't expect average folks to use it or
> develop for it?

Getting rid of blinkers?

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad

[Navin Johnson]

[-- Attachment #1: Type: text/plain, Size: 361 bytes --]

On 5/12/07, lucio@proxima.alt.za <lucio@proxima.alt.za> wrote:

>
> Getting rid of blinkers?


The long tail - "We gained more developers today that didn't write code at
all yesterday than we gained today of all the developers that did write
code yesterday". [1]

Yours faithfully,
Navin R. Johnson

[1] http://en.wikipedia.org/wiki/The_Long_Tail

[-- Attachment #2: Type: text/html, Size: 915 bytes --]

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad

[lucio]

> P.S. - This is my last posting with my real name. I'll use an alias from now on.

Please don't.  I have no idea who(m) I speak for, but I appreciate
comment that does not border on the whinge fringe.

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[Dave Lukes]

Vester Thacker wrote:
>  What is the purpose of writing an
>  operating system if you don't expect average folks to use it or
>  develop for it?

What is the point of developing a formula 1 racing car
if you don't expect average folks to use it or
produce aftermarket accessories for it?

I no more expect "average folks" (who are these, BTW?)
to use or develop for (or even notice the existence of) plan9
than I expect them to turn up at the shopping mall
in in an F1 Ferrari.

The following statements are provably untrue:
* Making something more popular makes it better.
* Making something better makes it more popular.
There are far too many counterexamples (Adolf Hitler, Betamax, ...).

So why do you care whether something is more popular?

Given a wonderful OS used by 10 people or a crap OS used by
10E6 people, I'll take the former.

D.


********************************************************************
This e-mail and files transmitted with it are confidential. If you 
have received this e-mail in error please notify us immediately by
return email or telephone +44 (0)207398 0207. You are not authorised
to, and must not, disclose, copy, distribute or retain this message
or any part of it.
********************************************************************

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[Navin Johnson]

On 5/16/07, Dave Lukes <davel@iontrading.com> wrote:
>
> So why do you care whether something is more popular?

If you have a cure, wouldn`t you`d like to share it? Why not provide
benefit to others?

This reminds me of "The Parable of the Excellent Physician and His
Sick Children".
http://www.nst.org/articles/Physician.txt

Yours faithfully,
Navin.r.j

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> If you have a cure, wouldn`t you`d like to share it? Why not provide
> benefit to others?

Not if the others are religious nuts who refuse treatment and are
likely to burn you at the stake for using infernal means to cure them.

That, by the way, is the reason why I will _not_ read the parable.

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

Dave Lukes <davel@iontrading.com> wrote:
>
> So why do you care whether something is more popular?

Actually, I do care.  The way I care in traffic that motorcycles are
not more popular and my life as a motorcyclist in greater danger
because of that.  But I do not care to make my and all other
motorcycles more car-like because of it.  It defeats all the
objectives of motorcycling in the first place.

And, yes, I rather like to think of Plan 9 as the hot-iron of the
OSes, the lean, mean machine.  The analogy is even better suited than
the Ferrari and/or F1 variety.

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Charles Forsyth]

> This reminds me of "The Parable of the Excellent Physician and His
> Sick Children".
	``Of the children who were more gravely poisoned, the 
	effects of the toxin were more severe and had consequently effected 
	their reasoning.''
well, surely that speaks for itself, but presumably the messenger
sent by the most excellent physician to tell his children
of his (feigned) death is Uriel? certainly he's always delivering that message!

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[Robert Sherwood]

[-- Attachment #1: Type: text/plain, Size: 181 bytes --]

More than that; hobbyists, however enlightened, cannot provide continuing
funding for Plan 9 development. Witness current hardware support. Plan 9 has
to grow or eventually die.

[-- Attachment #2: Type: text/html, Size: 183 bytes --]

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[W B Hacker]

Robert Sherwood wrote:
> More than that; hobbyists, however enlightened, cannot provide continuing
> funding for Plan 9 development. Witness current hardware support. Plan 9 
> has
> to grow or eventually die.
> 

The overall 'computerish' and network worlds are now large enough that even a 
very small niche can remain viable.

So I don't know that plan9 has to 'grow' as much as it has to be seen as 'of 
value' and still an environment where ideas are put into practice that are NOT 
done [well | at all] elsewhere. Much as is the case with a fair number of 
'embedded' OS.

IOW - it has to be seen as 'useful' to enough folks - be they students, 
researchers, purely experimenters, or pragmatists with specialized 'real world' 
(read 'commercial') applications not as well served with other tools.

Whether it [still | ever ] [does | did ] any/all of those things is a separate 
issue from mass popularity or 'growth'.

That said - I have not personally detected a particularly 'comfortable' level of 
  published evidence in that direction.

Even Vita Nuova's news releases seem few and far between, as if maintenance of a 
legacy base and a design-win every few *years* is all that is going on.

Bill

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[erik quanstrom]

	grow or die
this seems to me to be a statement, not an argument.

you also imply that hobbists can't write drivers.  does being
a hobbist set some bit in one's brain that inhibits the device-driver
region?

i suspect the real reason more devices are not supported is:
hardware makers often make it hard or impossible to
obtain specifications.  there are big exceptions to this.  ahci
was fully documented, for example.

with proper documentation there is nothing magic about a
device driver that makes it any more difficult to write than anything
else.

by the way, there are plenty of professionals working on plan 9.
(and at least 4 new drivers this year.)

- erik

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM ThinkPad R60e notebook compatible with Plan9?]

[Robert Sherwood]

[-- Attachment #1: Type: text/plain, Size: 1570 bytes --]

Two caveats:

1. I am not an expert in this area.
2. You're all probably sick of hearing these old arguments and should put me
in your kill file if my ignorance offends.

I've observed that real, consistent progress is largely made by paid
developers, not hobbyists. This is true even in open source projects. The
linux kernel, X windows, mozillla, and open office all have large, paid
development teams that do the bulk of the work. There are the occasional
hobbyist contributers that can provide real assistance, but the bulk is done
by paid developers. This is also true of plan 9, I believe.

In order to pay developers to develop, the paying organization has to see
some compelling benefit from the end result. For linux, mozilla, X and open
office, there are number of commercial entities that derive cost savings or
profit based on development of that software. Except mozilla. I'm not sure
why that's being funded. :)

Eventually, if Plan 9 can't demonstrate a unique business benefit. I.E.
someone can save significant money or make significant money by doing
something, but only if they use Plan 9, the paid developers will be sent to
other projects. That's the end of plan 9 for all practical purposes, if the
history of similar projects is any indication. Some day, even the most die
hard enthusiast will have to replace their creaky old computers with new
ones that have no driver support in plan 9.

I look forward to a heaping helping of scorn from young people on the list
who are insulated from commercial considerations :)

[-- Attachment #2: Type: text/html, Size: 1668 bytes --]

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> More than that; hobbyists, however enlightened, cannot provide continuing
> funding for Plan 9 development. Witness current hardware support. Plan 9 has
> to grow or eventually die.

Well, we've been warned.

Given the alternative between diluting Plan 9 to suit the demand for
snazz (who's going to deliver that, anyway?) and watching Plan 9
become irrelevant to the marketplace, I'll pick the latter any time.

As for the real alternative, which is for Plan 9 to become more
Linux-like which means more Windows-like, then what's the point?
Linux is there, Windows is there, why have a third contender?  What
innovation does Plan 9 contribute that the public is actually
clamouring for?

In fact, I'd hazard that Linux's only asset is its cost, in the eyes
of the consumers.  Sadly, no other OS can beat that cost.  Actually,
delete that "sadly".

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Charles Forsyth]

> Linux is there, Windows is there, why have a third contender?  What
> innovation does Plan 9 contribute that the public is actually
> clamouring for?

good sense of humour

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

>> Linux is there, Windows is there, why have a third contender?  What
>> innovation does Plan 9 contribute that the public is actually
>> clamouring for?
> 
> good sense of humour

Sorry, I don't qualify.  I have been accused of not having one.

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[W B Hacker]

lucio@proxima.alt.za wrote:
>> More than that; hobbyists, however enlightened, cannot provide continuing
>> funding for Plan 9 development. Witness current hardware support. Plan 9 has
>> to grow or eventually die.
> 
> Well, we've been warned.
> 
> Given the alternative between diluting Plan 9 to suit the demand for
> snazz (who's going to deliver that, anyway?) and watching Plan 9
> become irrelevant to the marketplace, I'll pick the latter any time.
> 

That doesn't really seem to be the choice. There is a middle ground.

'Current' drivers for networking would seem to be critical path, audio-visual not.

> As for the real alternative, which is for Plan 9 to become more
> Linux-like which means more Windows-like, then what's the point?
> Linux is there, Windows is there, why have a third contender?  What
> innovation does Plan 9 contribute that the public is actually
> clamouring for?

Think PDA, phone handset, 'thin client' (terminal) and the heavy-hitters for 
storage and computation located somewhere else on the network.

Sure - the need is being filled with WinCE, Palm, Symbian, even stripped-down 
Linux already.

But if ever there was a market born to take best advantage of Plan9's long suit, 
handheld, or 'wearable' has to be the most obvious contender, and on power nd 
bandwidth consumption as much as CPU cycles or 'local' RAM capacity.

> 
> In fact, I'd hazard that Linux's only asset is its cost, in the eyes
> of the consumers.  Sadly, no other OS can beat that cost.  Actually,
> delete that "sadly".
> 
> ++L
> 
> 

The *BSD's beat Linux 'cost' quite handily - even if CD's for both are 
purchased, not downloaded.

Linux rapid and 'diffused' devel model and plethora of 'distros' creates a need 
for for more time invested in migrating, porting, upgrading, seeking answers - 
retraining, 'er 'keeping current'.

Grant, a *BSD might not be the best choice for playing music, videos, or games 
(save perhaps OS X).

But OS X *also* beats Linux' cost, hands-down - and even on 50% to 100% more 
costly hardware - unless one values time at a *negative* per-hour figure.

None of which is all that relevant to what Plan9 is best at.

Sharing networked resources per se?  Not that *alone*.

Scitek/IBM/MS NETBIOS & SMB 'net use' or Novell 'attach' were there years 
earlier than 'bind', get much the same end-results.

So too other Xerox-derived contemporaries (VINES, StreetTalk, etc.). Even MAP/TOP.

But most of those are not as clean or efficient, let alone 'orthogonal' as the 
Plan9 model.

Nor are their communications necessarily as robust. 'Early' Netware the 
exception, when it still generally had an essentially 'no-fail' and 
deterministic network physical layer, i.e. ARCNET, TCNS, 100-VG-AnyLAN.


On technical merit, Plan9 *should* be making inroads into the networked mobile 
market. And Alcatel-Lucent *are* players there.

But too many folks are willing to either consider Plan9 effectively dead or 
would like to keep it in a coma so as to 'feel righteous'. скопцы - like.

Was it 'Glenda' that Willie Nelson was singing about?

"...And sometimes it seems ... that she ain't worth the trouble at all
But she could be worth the world ...if somehow you could touch her at all.."


Bill

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Uriel]

> Think PDA, phone handset, 'thin client' (terminal) and the heavy-hitters for
> storage and computation located somewhere else on the network.
>
> Sure - the need is being filled with WinCE, Palm, Symbian, even stripped-down
> Linux already.
>
> But if ever there was a market born to take best advantage of Plan9's long suit,
> handheld, or 'wearable' has to be the most obvious contender, and on power nd
> bandwidth consumption as much as CPU cycles or 'local' RAM capacity.

You have to be nuts to use Plan 9 for such things when you can use Inferno.

uriel

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[W B Hacker]

Uriel wrote:
>> Think PDA, phone handset, 'thin client' (terminal) and the 
>> heavy-hitters for
>> storage and computation located somewhere else on the network.
>>
>> Sure - the need is being filled with WinCE, Palm, Symbian, even 
>> stripped-down
>> Linux already.
>>
>> But if ever there was a market born to take best advantage of Plan9's 
>> long suit,
>> handheld, or 'wearable' has to be the most obvious contender, and on 
>> power nd
>> bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
> 
> You have to be nuts to use Plan 9 for such things when you can use Inferno.
> 
> uriel
> 

Trolling, are you?

Aside from their common roots 'native' Plan9 retains an efficiency edge over 
'native' Inferno & limbo.

Most Inferno installs seem to sit atop another full-size OS, and too-seldom is 
that OS Plan9, or even remotely similar. Efficiency is bound to suffer.

I do NOT like 'C' - but the interpreted language has not yet been born that can 
come close to matching compiled-C for speed of execution. Not to mention the 
massive weight of publically available prior art - and the 'artists' with 
experience to adapt and create.

And those do still matter.

Bill

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> Think PDA, phone handset, 'thin client' (terminal) and the heavy-hitters for 
> storage and computation located somewhere else on the network.

I do, but what induces the IT manager in a 200+ organisation to think
Plan 9?  And why would I want her to?

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[W B Hacker]

lucio@proxima.alt.za wrote:
>> Think PDA, phone handset, 'thin client' (terminal) and the heavy-hitters for 
>> storage and computation located somewhere else on the network.
> 
> I do, but what induces the IT manager in a 200+ organisation to think
> Plan 9?  And why would I want her to?
> 
> ++L
> 
> 

I'd class that as a de-facto impossibility.

WTH - we couldn't get OS/2 'sold' in an organization ten times that size and 
more.  Even though it was largely IBM mainframe dominated, and whose senior 
management *liked* IBM .... and against, of all things, Win 3.X - despite a more 
robust and then still fully compatible Win-OS2 being built in to OS/2. (Win-95 
was still years away at the time).

I don't see Plan9 ever even starting down the 'populist' road in that manner.

But the choice of embedded and 'appliance' OS'en is not made the same way, and 
the 'Windows' cachet as not as hard to compete with.

Virus host in my cellphone?  No thanks!

But cellphones, to name just one - can easily sell half-a-million 'seats' in two 
years or fewer - and to folks who could give a Massachusetts as to what brand is 
on the underlying OS software.

Those who have WinCE on their 'devices' know it. Those who have 'ABM' often 
*don't* know - or care - what they are using. So long as it JFW.

To the extent Plan9 can support a 'bespoke' UI and show life-cycle cost and 
performance advantages, STB, hand-held, and other 'appliance' makers can adopt 
it without it being directly 'in the face' of the end user.

And by 'bespoke UI' I don't mean folks here porting X-Windows, re-inventing 
drawterm or anything of the sort.

Rather, the device-maker's own team putting up something customized - more like 
GEM or the iPOD interface - even QNX' 'Photon' approach.

Something that 'JFDI' whatever the device is expected to do.

Not having a heavy and entrenched GUI already in place to 'fight with' is 
actually an advantage for Plan9 on that score.

Just do a file-count and 'du' before and after an install of 
Xorg+wm+desktop+tools of-choice on any *n*x.

Then weep over the waste of it all. And prepare for version Hell on upgrades.

There has to be a better way.

Bill

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> There has to be a better way.

There is a better way, but the decision makers are getting their
salaries by towing a line, not being innovative.  They have been
brain-washed into that frame of mind and they will not budge.  In the
meantime, the marketeers keep rasing the expectations and, as long as
Moore's Law is not shattered, the engineers keep delivering.

Only Moore can break that vicious circle and the likelyhood that it
will be in Plan 9's favour is minimal.  In fact, when the vicious
circle breaks, all bets are off.

++L

[9fans] Wearables

[john]

> lucio@proxima.alt.za wrote:
> But if ever there was a market born to take best advantage of Plan9's long suit, 
> handheld, or 'wearable' has to be the most obvious contender, and on power nd 
> bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
> 

A friend and I are starting a project to create a simple wearable computer. We've
got some hardware to get started; probably will begin with a laptop, our camera
viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or 
some form of home-brewed chording device. I considered using Plan 9, but since
we don't plan to include a pointing device yet, and the viewfinder can only display
low resolutions and in black and white, I think we'll end up going with something
designed to be used 80x24 characters at a time... Linux. If somebody can present
me with some good reasons to use Plan 9 instead, we can try it, but I really
don't think Plan 9 actually is ideal for a wearable.

John

Re: [9fans] Wearables

[Gabriel Diaz]

hello

import wearable_thing_in_front_of_me /n/friend
cp $home/images/friendly_foto.jpg /n/friend

:-?

may be you can speak with Nemo, i think the wearable thing is just the
same work they are doing with planb/plan-c?, so he probably has not
only ideas, but experiencies in doing such things.

slds.

gabi


On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> > lucio@proxima.alt.za wrote:
> > But if ever there was a market born to take best advantage of Plan9's long suit,
> > handheld, or 'wearable' has to be the most obvious contender, and on power nd
> > bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
> >
>
> A friend and I are starting a project to create a simple wearable computer. We've
> got some hardware to get started; probably will begin with a laptop, our camera
> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or
> some form of home-brewed chording device. I considered using Plan 9, but since
> we don't plan to include a pointing device yet, and the viewfinder can only display
> low resolutions and in black and white, I think we'll end up going with something
> designed to be used 80x24 characters at a time... Linux. If somebody can present
> me with some good reasons to use Plan 9 instead, we can try it, but I really
> don't think Plan 9 actually is ideal for a wearable.
>
> John
>
>

Re: [9fans] Wearables

[David Leimbach]

On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> > lucio@proxima.alt.za wrote:
> > But if ever there was a market born to take best advantage of Plan9's long suit,
> > handheld, or 'wearable' has to be the most obvious contender, and on power nd
> > bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
> >
>
> A friend and I are starting a project to create a simple wearable computer. We've
> got some hardware to get started; probably will begin with a laptop, our camera
> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or
> some form of home-brewed chording device. I considered using Plan 9, but since
> we don't plan to include a pointing device yet, and the viewfinder can only display
> low resolutions and in black and white, I think we'll end up going with something
> designed to be used 80x24 characters at a time... Linux. If somebody can present
> me with some good reasons to use Plan 9 instead, we can try it, but I really
> don't think Plan 9 actually is ideal for a wearable.
>

What are your requirements that disqualify Plan 9?  Or is just a "feeling"?

Coraid puts Plan 9 in their storage products.  It's performing a
useful task, and you don't even need to know it's there.  But it's
still powering the device.

Small and simple can still be beautiful and elegant :-)

> John
>
>


-- 
- Passage Matthew 5:37:
   But let your communication be, Yea, yea; Nay, nay: for whatsoever
is more than these cometh of evil.

Re: [9fans] Wearables

[john]

> On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
>> > lucio@proxima.alt.za wrote:
>> > But if ever there was a market born to take best advantage of Plan9's long suit,
>> > handheld, or 'wearable' has to be the most obvious contender, and on power nd
>> > bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
>> >
>>
>> A friend and I are starting a project to create a simple wearable computer. We've
>> got some hardware to get started; probably will begin with a laptop, our camera
>> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or
>> some form of home-brewed chording device. I considered using Plan 9, but since
>> we don't plan to include a pointing device yet, and the viewfinder can only display
>> low resolutions and in black and white, I think we'll end up going with something
>> designed to be used 80x24 characters at a time... Linux. If somebody can present
>> me with some good reasons to use Plan 9 instead, we can try it, but I really
>> don't think Plan 9 actually is ideal for a wearable.
>>
> 
> What are your requirements that disqualify Plan 9?  Or is just a "feeling"?
> 
> Coraid puts Plan 9 in their storage products.  It's performing a
> useful task, and you don't even need to know it's there.  But it's
> still powering the device.
> 
> Small and simple can still be beautiful and elegant :-)
> 

You read the thing, right? I said that, with our display device, it's probably
only feasible to work at a terminal. We connected the display to a Windows
machine to test it. You can read a standard 80x24 terminal, but barely.

Plan 9's support for just plain, old, no-windowing-system terminal
use is pretty much nonexistent, so there's a good reason.

Yes, we could cook up a lib/profile that automatically starts a big rc window,
but you still need to use a pointing device of some kind. Yes, we can strap on
a trackball, but I'd like to try reducing the amount of equipment hanging from
my belt. If somebody would like to donate a Twiddler keyboard, then I'd be all
for trying Plan 9.

John

Re: [9fans] Wearables

[W B Hacker]

john@csplan9.rit.edu wrote:
>> lucio@proxima.alt.za wrote:
>> But if ever there was a market born to take best advantage of Plan9's long suit, 
>> handheld, or 'wearable' has to be the most obvious contender, and on power nd 
>> bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
>>
> 
> A friend and I are starting a project to create a simple wearable computer. We've
> got some hardware to get started; probably will begin with a laptop, our camera
> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or 
> some form of home-brewed chording device. I considered using Plan 9, but since
> we don't plan to include a pointing device yet, and the viewfinder can only display
> low resolutions and in black and white, I think we'll end up going with something
> designed to be used 80x24 characters at a time... Linux. If somebody can present
> me with some good reasons to use Plan 9 instead, we can try it, but I really
> don't think Plan 9 actually is ideal for a wearable.
> 
> John
> 
> 

'Ideal' only in two senses:

- Very well-suited to having the 'heavy' resources remoted over reasonably 
efficient (low bandwidth) networking.

- lacking a GP GUI (rio/acme are, IMNSHO, a coder's IDE, not a GP GUI), but 
having lightweight tools to implement one (drawterm, VNC)  - so you can do 
'locally' only what your app really *must* do locally.

As to 'pointing device' - why not a tilt-disk, 'clit' or trackball? All of which 
are cheaply salvaged from new or used hardware. Chording the 'Plan9 way' is not 
an absolute requirement - just one already built-in.

Viewing device?  'Virtual reality' headset, perhaps?

Or go the other way...

text-to-speech in an earpiece, speech-to-text from a mic.

'Heavy' CPU to convert bothways accurately is remoted.

Might mean the heaviest thing you have to wear is...

...a 'dumb' telephone handset and a thin LCD for graphics when needed.

My biggest personal objection to most modern PDA/phone rigs (Blackberry, Treo, 
et al) is the need to grab a stylus and/or otherwise use BOTH hands when NO 
hands is a nicer goal, and ONE hand was possible even with the ancient HP-200-LX 
(thumb-typing).

Belt-mount and Bluetooth or similar seems a good idea though.

Linux? Far too 'heavy', even stripped - which is not as easy as it sounds if you 
need even basic functionality). if not Plan9, then Minix3 revanche is lighter 
(and very Posix compliant)

But might be better-off with DRDOS and GEM. Seriously.

Find an HP-100/200-LX (MSDOS, not DRDOS) and see what was possible lo those many 
years ago with a couple of the right PCMCIA cards and lithium AA batteries.

Used to carry a pair of clip leads and external twin D-cell holder to send faxes 
and login to CompuServe from hotel rooms. Purchased and discarded batteries 
locally so as to not have to carry the weight or a charger. ELSE 'borrowed' the 
rechargeable emergency flashlight found in many hotels.

'Too soon we forget' how much could be accomplished with a lowly VT-whatever 
'dumb terminal' connected to the right support infrastructure at a mere 1200 - 
9600 bps.

These need not replace the entire laptop/desktop 'puterish experience - just 
bridge the gaps.

Bill

Re: [9fans] Wearables

[john]

> john@csplan9.rit.edu wrote:
>>> lucio@proxima.alt.za wrote:
>>> But if ever there was a market born to take best advantage of Plan9's long suit, 
>>> handheld, or 'wearable' has to be the most obvious contender, and on power nd 
>>> bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
>>>
>> 
>> A friend and I are starting a project to create a simple wearable computer. We've
>> got some hardware to get started; probably will begin with a laptop, our camera
>> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or 
>> some form of home-brewed chording device. I considered using Plan 9, but since
>> we don't plan to include a pointing device yet, and the viewfinder can only display
>> low resolutions and in black and white, I think we'll end up going with something
>> designed to be used 80x24 characters at a time... Linux. If somebody can present
>> me with some good reasons to use Plan 9 instead, we can try it, but I really
>> don't think Plan 9 actually is ideal for a wearable.
>> 
>> John
>> 
>> 
> 
> 'Ideal' only in two senses:
> 
> - Very well-suited to having the 'heavy' resources remoted over reasonably 
> efficient (low bandwidth) networking.
> 
> - lacking a GP GUI (rio/acme are, IMNSHO, a coder's IDE, not a GP GUI), but 
> having lightweight tools to implement one (drawterm, VNC)  - so you can do 
> 'locally' only what your app really *must* do locally.

I'm not really looking forward to dealing with rio at the very low resolutions
our device uses. The idea of window management on such a display, barring
the use of wmii or dwm, seems ludicrous.


> As to 'pointing device' - why not a tilt-disk, 'clit' or trackball? All of which 
> are cheaply salvaged from new or used hardware. Chording the 'Plan9 way' is not 
> an absolute requirement - just one already built-in.
> 
> Viewing device?  'Virtual reality' headset, perhaps?

As I mentioned initially, we're using a camcorder viewfinder as the display.
That means black and white, low resolution (composite input). Until we can
get club approval (Robotics club), our budget is nil; we're working on
scavenged equipment.

> 
> Or go the other way...
> 
> text-to-speech in an earpiece, speech-to-text from a mic.
> 
> 'Heavy' CPU to convert bothways accurately is remoted.
> 
> Might mean the heaviest thing you have to wear is...
> 
> ...a 'dumb' telephone handset and a thin LCD for graphics when needed.

Although we do intend to write specialized code for the project, I do not
want to write a text-to-speech and speech-to-text suite right now.


> 
> My biggest personal objection to most modern PDA/phone rigs (Blackberry, Treo, 
> et al) is the need to grab a stylus and/or otherwise use BOTH hands when NO 
> hands is a nicer goal, and ONE hand was possible even with the ancient HP-200-LX 
> (thumb-typing).
> 
> Belt-mount and Bluetooth or similar seems a good idea though.
> 
> Linux? Far too 'heavy', even stripped - which is not as easy as it sounds if you 
> need even basic functionality). if not Plan9, then Minix3 revanche is lighter 
> (and very Posix compliant)

Most likely candidate right now is a Linux laptop in a shoulder bag or Slackware
installed on a mini PC we have in the lab.
 
> But might be better-off with DRDOS and GEM. Seriously.
> 
> Find an HP-100/200-LX (MSDOS, not DRDOS) and see what was possible lo those many 
> years ago with a couple of the right PCMCIA cards and lithium AA batteries.
> 
> Used to carry a pair of clip leads and external twin D-cell holder to send faxes 
> and login to CompuServe from hotel rooms. Purchased and discarded batteries 
> locally so as to not have to carry the weight or a charger. ELSE 'borrowed' the 
> rechargeable emergency flashlight found in many hotels.
> 
> 'Too soon we forget' how much could be accomplished with a lowly VT-whatever 
> 'dumb terminal' connected to the right support infrastructure at a mere 1200 - 
> 9600 bps.
> 

This is what I want. That's why Linux seems like a decent candidate. The hardware
we're looking at can handle it just fine. It gives us a VT type interface. Couple
that with campus wireless and we can connect to whatever "support infrastructure"
we want.


John "Plan 9 Koolaid" Floren

Re: [9fans] Wearables

[W B Hacker]

john@csplan9.rit.edu wrote:
*snip*

> 
> Although we do intend to write specialized code for the project, I do not
> want to write a text-to-speech and speech-to-text suite right now.
>

You don't need to write it. As with AT&T's famous cross of a Rooster and a 
telephone pole, you just need to 'reach out and touch' it.

Networking the audio bothways is the equvalent of the rest of that joke.

*snip*

> 
> Most likely candidate right now is a Linux laptop in a shoulder bag or Slackware
> installed on a mini PC we have in the lab.
>

Seems driven by 'familiarity' rather than 'suitability for the purpose'.

>> But might be better-off with DRDOS and GEM. Seriously.
>>

Or Forth. I've done ISAM DB's, multi-tasking, high-speed terminals, graphical 
games, editors, text to speech, and a lot more in it - always in well under 24 KB.

As the language is also a virtual-memory OS, and is *tiny*, it could get into 
and STAY in the L1 cache of most modern CPU.

There are others like that.

The trick is to NOT otherwise carry a full-blown 'OS' such as Linux in order to 
support the 'applications', but rather to provide the applications with only the 
I/O hardware layer interface they will actually use.

*snip*

> 
> This is what I want. That's why Linux seems like a decent candidate. The hardware
> we're looking at can handle it just fine. It gives us a VT type interface.

CP/M gives a 'VT like interface' in a couple of KB. Real VT's. ADM-3's SOroc IQ 
120's TVi 920's et al did it with gate arrays.

Another 500 bytes and you have serial port and Hayes-compatible modem drivers. 
Not a whole lot more was needed for David Clark's IP stack.

Linux is a 'decent candidate' the way delivering pizza in a motorhome compares 
to using a Vespa scooter.

 > Couple
> that with campus wireless and we can connect to whatever "support infrastructure"
> we want.
>

Familiarity with Linux seems to shut down common-sense somehow.

A friend in Hong Kong spent two years working on a contract for a vending 
machine maker.  He was trying to develop an embedded Linux variant to automate 
vending machine inventory - to be 'polled' from HQ via an embedded GSM device.

Meanwhile. KISS.

Back on Planet Earth, someone else hooked an ignorant Programmed Logic Array to 
full/empty slot sensors, simply shifted the current count out as a number when 
the phone module in the vending machine was polled. No need for even a TOD 
function. No 'OS'. No UPS. No need even for NVRAM - just re-poll the sensors 
when called.

HQ knows the time of day and can keep their own copy of inventory.

Same again with the now-prevalent smart-card payment modules. Move the data, 
move the work needed to manipulate it to where reseources are cheapest and most 
easily maintained.

'When the only tool you have is an Operating System...'

it is hard to remember that you only have a few specific taska to accomplish - 
'state machine work' most of them,

..and that 'general purpose' programmability is better served back at the 
Mothership or elsewhere on the network - not by over-empowering the 
should-have-been-thin client/dumb device.

All that does is make money for the battery makers and chiropractors.

Plan9 took cognizance of that, and it isn't just about whether you *can* get 
enough power into a postage-stamp size device, or even how cheaply.

To an extent, it is about where you want to do your backups, maintenance 
upgrades, devel, testing, and support for the least cost and hassle *those* entail.

Bill

Re: [9fans] Wearables

[ron minnich]

On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> > lucio@proxima.alt.za wrote:
> > But if ever there was a market born to take best advantage of Plan9's long suit,
> > handheld, or 'wearable' has to be the most obvious contender, and on power nd
> > bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
> >
>
> A friend and I are starting a project to create a simple wearable computer. We've
> got some hardware to get started; probably will begin with a laptop, our camera
> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or
> some form of home-brewed chording device. I considered using Plan 9, but since
> we don't plan to include a pointing device yet, and the viewfinder can only display
> low resolutions and in black and white, I think we'll end up going with something
> designed to be used 80x24 characters at a time... Linux. If somebody can present
> me with some good reasons to use Plan 9 instead, we can try it, but I really
> don't think Plan 9 actually is ideal for a wearable.

80x24 eh? So, the provision, or lack thereof, of a "glass tty" is the
deciding factor?

I think your decision tree needs to have a few branches grafted on :-)

ron
p.s. Due to new workplace health and safety regulations, I am not
allowed to tell you how to pronounce "glass tty"

Re: [9fans] Wearables

[john]

> On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
>> > lucio@proxima.alt.za wrote:
>> > But if ever there was a market born to take best advantage of Plan9's long suit,
>> > handheld, or 'wearable' has to be the most obvious contender, and on power nd
>> > bandwidth consumption as much as CPU cycles or 'local' RAM capacity.
>> >
>>
>> A friend and I are starting a project to create a simple wearable computer. We've
>> got some hardware to get started; probably will begin with a laptop, our camera
>> viewfinder HMD, and a keyboard strapped around the waist (crude, I know) or
>> some form of home-brewed chording device. I considered using Plan 9, but since
>> we don't plan to include a pointing device yet, and the viewfinder can only display
>> low resolutions and in black and white, I think we'll end up going with something
>> designed to be used 80x24 characters at a time... Linux. If somebody can present
>> me with some good reasons to use Plan 9 instead, we can try it, but I really
>> don't think Plan 9 actually is ideal for a wearable.
> 
> 80x24 eh? So, the provision, or lack thereof, of a "glass tty" is the
> deciding factor?
> 
> I think your decision tree needs to have a few branches grafted on :-)
> 
> ron
> p.s. Due to new workplace health and safety regulations, I am not
> allowed to tell you how to pronounce "glass tty"

Well, Ron, as explained elsewhere, our display gets hard to read pretty easily.
Even 80x24 may be too small for continued use; it's something that will need to
be tested. If you or anybody else wants to send us a better head-mounted
display, do it. For now, I'm going to concentrate on using the (free) display
we have... people were apparently productive for 40+ years using only TTYs,
glass or paper as it may be.

John

Re: [9fans] Wearables

[Charles Forsyth]

i didn't really understand why being limited to 24x80 (or whatever) was
an inherent block to the use of plan 9.  indeed you wouldn't use
rio, but that's just a program and you can run another.

Re: [9fans] Wearables

[ron minnich]

On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:

> Well, Ron, as explained elsewhere, our display gets hard to read pretty easily.
> Even 80x24 may be too small for continued use; it's something that will need to
> be tested.

I'm with Charles on this one. I do a lot of work with Plan 9 systems
that don't use rio or any wm for that matter. On those systems, the
network is everything. Maybe you're not going to run connected.

But, in our case, we had systems w/o a graphical display, that I
needed to run connected, and I would have been very unhappy doing
those systems with anything but Plan 9.

I see your point re the display, but based on our experience, I still
think Linux is going to limit you in the end.

That said, there's lots more tty eye candy for Linux these days ...
some things are truly easier with Linux. Just none of the network
things.

thanks

ron

Re: [9fans] Wearables

[john]

> On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> 
>> Well, Ron, as explained elsewhere, our display gets hard to read pretty easily.
>> Even 80x24 may be too small for continued use; it's something that will need to
>> be tested.
> 
> I'm with Charles on this one. I do a lot of work with Plan 9 systems
> that don't use rio or any wm for that matter. On those systems, the
> network is everything. Maybe you're not going to run connected.
> 
> But, in our case, we had systems w/o a graphical display, that I
> needed to run connected, and I would have been very unhappy doing
> those systems with anything but Plan 9.
> 
> I see your point re the display, but based on our experience, I still
> think Linux is going to limit you in the end.
> 
> That said, there's lots more tty eye candy for Linux these days ...
> some things are truly easier with Linux. Just none of the network
> things.
> 
> thanks
> 
> ron

Plan 9 fails for GUI-less use. If we go to a GUI system at some point,
I think it may be back in the running, but consider the current 
requirements. We need something:
-That has a good CLI
-That can handle wireless
-That has a lot of CLI-oriented applications

Linux has support for a *lot* of wireless cards. It's based on a system
that was designed for CLI use. It's got emacs, so I guess that answers
all questions about CLI-oriented applications.
Plan 9... working at the plain old command line, can you even interrupt
a program? The two main editors are GUI based (not gonna use 'ed'). As
seems apparent, it has far fewer supported wireless cards. It runs on
fewer machines.
Somebody, quick, send me a decent head mounted display, a Twiddler,
and a little machine that can run Plan 9 with supported wireless. I'll
set it all up, use it to access csplan9. Then I'll post pictures on the
wiki. Until then, I guess I'm still gonna end with Linux.

John 

Re: [9fans] Wearables

[Skip Tavakkolian]

> Until then, I guess I'm still gonna end with Linux.

cool!  good luck.

Re: [9fans] Wearables

[W B Hacker]

Skip Tavakkolian wrote:
>> Until then, I guess I'm still gonna end with Linux.
> 
> cool!  good luck.
> 
> 

'....end with Linux'.

Perhaps Wyeth Consumer Healthcare might be interested in such 'out of sight ' 
wearables. No shortage of openings...

;-)


Bill

Re: [9fans] Wearables

[Salva Peiró]

On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> Plan 9 fails for GUI-less use. If we go to a GUI system at some point,
> I think it may be back in the running, but consider the current
> requirements. We need something:
> -That has a good CLI
> -That can handle wireless
> -That has a lot of CLI-oriented applications
>
> Linux has support for a *lot* of wireless cards. It's based on a system
> that was designed for CLI use. It's got emacs, so I guess that answers
> all questions about CLI-oriented applications.
> Plan 9... working at the plain old command line, can you even interrupt
> a program? The two main editors are GUI based (not gonna use 'ed'). As
> seems apparent, it has far fewer supported wireless cards. It runs on
> fewer machines.
> Somebody, quick, send me a decent head mounted display, a Twiddler,
> and a little machine that can run Plan 9 with supported wireless. I'll
> set it all up, use it to access csplan9. Then I'll post pictures on the
> wiki. Until then, I guess I'm still gonna end with Linux.
>
> John
>
>

Although this it slighty diverts from wearable to pocket-able,
I've used Inferno emu over Linux on a handheld [1] for 6 months now,
and i've find it comfortable enought, to continue using it.
(and recently it's got bt network support, even i've not used it yet)
So i would consider this another alternative.

[1] http://www.caerwyn.com/ipn/2006/12/inferno-emu-on-palm-tungsten-t3.html

-- 
salva

Re: [9fans] Wearables

[erik quanstrom]

On Thu May 17 12:40:05 EDT 2007, john@csplan9.rit.edu wrote:

> It's got emacs, so I guess that answers
> all questions about CLI-oriented applications.

oh yes it does.  like the sound of one hand clapping.

this really has nothing to do with plan 9.  perhaps this
discussion should be moved to alt.os.emacs.

- erik

Re: [9fans] Wearables

[ron minnich]

On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:

> Plan 9 fails for GUI-less use.

Plan 9 fails at anything that it doesn't succeed at.

This is the Plan 9 cycle of existence:

Since it's not good at X, people don't make it good at X. So they
don't use it for X. Since they don't use it for X, it's not good at X.
Since it's not good at X ..

It reminds me of the recycling mobius strip.

There's only way one out of this cycle, but it's painful.

Plan 9 CLI interface is, right now, where the Unix CLI was when I
started using it -- minus, of course, DEL. That's pretty easy to fix
-- well, trivial, in fact, to fix. In fact, someone I know has fixed
it. Solution left to reader. It's not a character builder, as it is
too easy.

Is there a fundamental reason that Plan 9 can not be a CLI system like
Unix? No.
Is anyone going to do it? Guess not.
If you had a decent Plan 9 CLI system, would it be nicer for these
network widgets than Linux? My experience says yes, because even an
indecent plan 9 sytsem is better for wireless network widgets.

ron

Re: [9fans] Wearables

[tlaronde]

On Thu, May 17, 2007 at 10:18:43AM -0700, ron minnich wrote:
> On 5/17/07, john@csplan9.rit.edu <john@csplan9.rit.edu> wrote:
> 
> >Plan 9 fails for GUI-less use.
> 
>[...]
> Plan 9 CLI interface is, right now, where the Unix CLI was when I
> started using it -- minus, of course, DEL. That's pretty easy to fix
> -- well, trivial, in fact, to fix. In fact, someone I know has fixed
> it. Solution left to reader. It's not a character builder, as it is
> too easy.
> 
> Is there a fundamental reason that Plan 9 can not be a CLI system like
> Unix? No.
> Is anyone going to do it? Guess not.

Since I will need it for KerGIS (the BSD licenced revival of the CERL's
GIS called GRASS), it may happen someday ;)

KerGIS uses range from handhelds (for data taken on the ground) to huge
computing power needs for some treatments.

Every problem I stumbled upon and about which I thought: it could be
great if... Plan9 has an answer, or can be extended to have one.

Look at the amount of code of GPLed GRASS. Look at the amount of code of
the BSD version (KerGIS). I started alone.
I'm still alone. But things are done. Slowly for some parts (there are
priorities), but done. If one wants, one can ;)
-- 
Thierry Laronde (Alceste) <tlaronde +AT+ polynum +dot+ com>
                 http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

Re: [9fans] Wearables

[Charles Forsyth]

> Plan 9... working at the plain old command line, can you even interrupt
> a program? The two main editors are GUI based (not gonna use 'ed'). As

so, you think that you can interrupt a program in rio because it uses draw?

if linux is such an answer to your prayers, as you say, i'm not quite sure why
you bothered to post to this list.

Re: [9fans] Wearables

[Charles Forsyth]

> i'm not quite sure why you bothered to post to this list.

i know: you're trying to make us jealous,
but we've already got one ...

	GUARD:  Oh, yes, it's very nice-a [To Other Guards]  I told him we already got one.

Re: [9fans] Wearables

[lucio]

>> lucio@proxima.alt.za wrote:
>> But if ever there was a market born to take best advantage of Plan9's long suit, 

Misquote.  The above is Bill Hacker's reply to my part in the rant
(for the record, no offence taken).

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Cranky Old Bat]

On Wed, 16 May 2007 18:26:12 GMT
lucio@proxima.alt.za wrote:

> 
> As for the real alternative, which is for Plan 9 to become more
> Linux-like which means more Windows-like

I see people say this all the time here but it makes no sense -- unless
maybe those who say it have only ever seen linux with KDE?
I'm typing this in sylpheed running in ion3. Doesn't look or feel a bit
like windows to me...
Most of my work is done from the command line.
How exactly do you believe this is windows-like?

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> Most of my work is done from the command line.
> How exactly do you believe this is windows-like?

So you're representative of the majority of Linux users?

Windows has a command line too.  Does that make it Unix-like?

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Devon H. O'Dell]

i know a thread full of people who are going to my spam folder.

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[David Leimbach]

On 5/18/07, Devon H. O'Dell <devon.odell@gmail.com> wrote:
> i know a thread full of people who are going to my spam folder.
>

So to combat the length of this thread you decided to add to it?

It's like the people who read tabloids who feel badly for Britney
Spears cuz she had a nervous breakdown from all the tabloid
attention...

Best way to get rid of threads like this is to ignore them out of existence.

Which I just failed to do.  DAMNIT!

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[ron minnich]

What do you have against Britney?

ron

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Skip Tavakkolian]

> What do you have against Britney?
> 
> ron

oh-oh, she did it again!

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Enrico Weigelt]

* ron minnich <rminnich@gmail.com> wrote:
> What do you have against Britney?

Today, she looks like she had too much apple pie ;-O


cu
-- 
---------------------------------------------------------------------
 Enrico Weigelt    ==   metux IT service - http://www.metux.de/
---------------------------------------------------------------------
 Please visit the OpenSource QM Taskforce:
 	http://wiki.metux.de/public/OpenSource_QM_Taskforce
 Patches / Fixes for a lot dozens of packages in dozens of versions:
	http://patches.metux.de/
---------------------------------------------------------------------

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Markus Sonderegger]

[-- Attachment #1: Type: text/plain, Size: 96 bytes --]

> How exactly do you believe this is windows-like?

Because it's the same bloated crap code.

[-- Attachment #2: Type: message/rfc822, Size: 5851 bytes --]

From: Cranky Old Bat <indulekha.devi.dasi@gmail.com>
To: 9fans@cse.psu.edu
Subject: Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM
Date: Fri, 18 May 2007 09:54:09 -0700
Message-ID: <20070518095409.57508b85.cranky.old.bat@gmail.com>

On Wed, 16 May 2007 18:26:12 GMT
lucio@proxima.alt.za wrote:

> 
> As for the real alternative, which is for Plan 9 to become more
> Linux-like which means more Windows-like

I see people say this all the time here but it makes no sense -- unless
maybe those who say it have only ever seen linux with KDE?
I'm typing this in sylpheed running in ion3. Doesn't look or feel a bit
like windows to me...
Most of my work is done from the command line.
How exactly do you believe this is windows-like?

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Douglas A. Gwyn]

Dave Lukes wrote:
> Given a wonderful OS used by 10 people or a crap OS used by
> 10E6 people, I'll take the former.

The problem with that is that, except perhaps for research purposes,
an OS is not an end in itself, but rather a platform supporting
applications, which are what actually satisfy human needs.
As time goes on, the popular platforms acquire a large number
of apps that become ever more essential (PDF reader or MPG viewer,
for example).  If the OS developer population is below some
critical mass, it can't keep up with such user requirements and
eventually the platform becomes in effect unusable for what have
become everyday needs.  (There are similar problems in keeping up
with device driver support for new hardware.)  Linux seems to
have reached the critical threshold, so it is "alive and well";
Plan9 seems to be well below that threshold, and has become
largely irrelevant, except perhaps as a testing ground for ideas
that may get adopted into more popular platforms.

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Dave Lukes]

Douglas A. Gwyn wrote:
>  Dave Lukes wrote:
> > Given a wonderful OS used by 10 people or a crap OS used by 10E6
> > people, I'll take the former.

>  The problem with that is that, except perhaps for research purposes,
>  an OS is not an end in itself, but rather a platform supporting
>  applications, which are what actually satisfy human needs.

To be specific:
    the _percieved_ _needs_ of _it's_ user base.

 > As time goes on, the popular platforms acquire a large number of apps 
that
>  become ever more essential

"Essential"?
AFAIK the list of human "essentials" still only includes
air, water, food, shelter and sex*.
All the rest is window dressing.

The above may sound facetious but is seriously intended:
what your user community views as essential is down to that community,
not down to some perceived need for global conformity.

>  (PDF reader or MPG viewer, for example).
Why does one _need_ an mpg viewer?

>  If the OS developer population is below some critical mass, it can't
>  keep up with such user requirements and eventually the platform
>  becomes in effect unusable for what have become everyday needs.

I refer to my formula-1 analogy.
a Ferrari _is_ "in effect unusable for ... everyday needs",
yet I hear no-one decrying the imminent demise of the brand.

>  (There are similar problems in keeping up with device driver support
>  for new hardware.)

Again, by analogy, F-1 teams don't suffer from this problem.
i.e. there are solutions to this, but they involve lots of money ....

 >   Linux seems to have reached the critical
>  threshold, so it is "alive and well"; Plan9 seems to be well below
>  that threshold, and has become largely irrelevant, except perhaps as
>  a testing ground for ideas that may get adopted into more popular
>  platforms.

Hmmm ... Bit like a formula-1 car, really ...

D.


* and chocolate or course.

********************************************************************
This e-mail and files transmitted with it are confidential. If you 
have received this e-mail in error please notify us immediately by
return email or telephone +44 (0)207398 0207. You are not authorised
to, and must not, disclose, copy, distribute or retain this message
or any part of it.
********************************************************************

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[lucio]

> * and chocolate or course.

Chocolate is part of sex.

++L

Re: Warning: Rant. Please disregard. [Was: Re: [9fans] Is IBM

[Gorka Guardiola]

On 5/17/07, Dave Lukes <davel@iontrading.com> wrote:
> "Essential"?
> AFAIK the list of human "essentials" still only includes
> air, water, food, shelter and sex*.
> All the rest is window dressing.
>
> >  (PDF reader or MPG viewer, for example).
> Why does one _need_ an mpg viewer?
>

For sex.
-- 
- curiosity sKilled the cat