Linux-Advocacy Digest #610, Volume #28 Thu, 24 Aug 00 04:13:04 EDT
Contents:
Re: The Powerful Linux GUI Interfaces - Re: Why is "ease of use" a dirty concept?
(R.E.Ballard ( Rex Ballard ))
----------------------------------------------------------------------------
From: R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]>
Subject: Re: The Powerful Linux GUI Interfaces - Re: Why is "ease of use" a dirty
concept?
Date: Thu, 24 Aug 2000 07:10:52 GMT
In article <[EMAIL PROTECTED]>,
"1$worth" <"1$worth"@costreduction.plseremove.screaming.net> wrote:
> s"R.E.Ballard ( Rex Ballard )" wrote:
> >
> > In article
<[EMAIL PROTECTED]>,
> > "1$Worth" <"1$Worth"@costreduction.plseremove.screaming.net>
wrote:
> [snip]
> > Actually, Linux has a number of GUIs, which, when the user is given
> > the choice of the "Redmond95" (windows-like) and others such as
> > KDE, Enlightenment, or even FVWM2 with AfterStep, indicates that
> > those interested in Linux are least interested in Redmond95 (the
> > look and feel of Microsoft).
>
> ..and this is a great strength.
Yes. The GUI is very much a matter of taste, of intangible preferences.
Some people love the Mac UI, others loved Windows 3.1, others loved
OS/2, and others loved Windows 95. Windows 98 even gives you a choice
of GUIs.
It's like body style. Some people like pick-ups, others like coups,
others like sedans, others like compacts, others like hatchbacks, and
still others like Luxury, Minivan, or SUV styles.
At one time, Ford only made the Model T in one body style. Very
quickly,
local machinists created a whole industry around restyling the bodies
of the "Washtub Model T". Many of the carriage and coach manufacturers
shifted to making bodies for Automobiles.
> > > not some ease of use BS [but to address the M$ rebooting idiom]"
> >
> > > For Linux to be propelled onto the desktop I'd urge
> > > posters to drop this assumption for the good of the community.
> >
> > I would agree. I'd also suggest that any Windows adovocate who
> > claimes that Linux lacks "ease of use" is speaking from a position
> > of total ignorance.
>
> I would disagree. Getting some hardware to work is a _real_ pain.
This is an installation issue. This is a very different issue that
IS being addressed by Manufacturers, OEMs, and VARs. It is an
interesting
discussion, and has been given it's own, dedicated newsgroup.
Comp.os.linux.setup covers these issues in excruciating detail.
If people don't want to fight through an installation, they should
either
get a machine preconfigured for Linux, or they should get a machine
that his been certified for Linux, or they should know that the
components
are on the compatibility list.
Linux is developing better and better USB support, they support
DVD-ROMs,
(but DVD-Movies are on-hold until DeCSS is resolved).
> This is not the fault of GUI of course, but to most people
> the GUI _is Linux
> and the configuration tools are not what they should be at this point
The current GUI tools available on Mandrake are quite competitive
with Windows' control panel. In fact, many of the configuration tools
are so similar to the Windows versions that I'm suprised Microsoft
hasn't raised "Look and Feel" issues.
> (even if improving all the time). Ease of use therefore must include
> making things easy for people - the whole point of the parc research.
Apple solved the problem by restricting the hardware supported.
They created the Mac, then the Mac II, then the Mac SE. The
only external interfaces on those machines were the SCSI port,
the proprietary Mac Printer port (to prevent the installation of
3rd party printers), and the Mac serial port. Even the mouse and
keyboard were Apple Exclusive.
Sun did the same thing with SunOS and Solaris. Eventually Solaris
was enhanced to support a variety of PC configurations, but even this
was based on techniques derived from Linux.
Ironically, even Microsoft's "plug-n-play" was inspired and driven
by Linux. Linux had plug an play in 1994 and Microsoft decided to
delay the release of "Chicago" (Windows 95) was originally to come
out in early 1995, and was moved to the last week of Microsoft's fiscal
1995 year. The SEC investigated the accounting and reporting practices
and found that Microsoft had declared pre-sold inventory shipped to
stores, and OEM initial purchase agreements as part of their 1995
income.
Microsoft knew how Linux did PnP and used NDAs to prevent Linux from
deriving PCI PnP. Because the PnP device simply identified the kind
of thing it was, the IHV still had to provide the available
configuration
options, and was prevented from disclosing them to the Linux community
by
NDAs.
Linux has done a pretty good job of dealing with PnP devices. USB is
the next big challenge. Again, for trivial devices such as Mice,
Keyboards, and Modems, Linux is pretty good at sorting it out. For
things like 250 meg zip drives, USB to Ethernet devices, and Scanners,
Cameras, and Video Capture units, some manufacturers are being very
Linux-friendly, others are too dependent on Microsoft's good will to
even offer binary drivers for Linux.
> >The fact is that many of the interfaces available
> > for Linux have been adapted from, and often enhanced by, some of the
> > best GUI designers in the industry, including members of PARC,
members
> > of NeXT, members of OpenLook, members of GNOME, and members of the
> > KDE projects. In addition, there will be new interfaces coming out
> > from the designers of the Apple Mac and iMac interfaces.
Furthermore,
> > Linux offers ways to get most of the best features of all of these
> > different environments.
>
> Hmmmm. You know and I know that integration is a real issue. Simple
> tasks such as drag-and-drop with copy/paste is not consistent and very
> off-putting for the user.
They are different, and in some cases handled very differently from
the way Microsoft does things. For example, if you mark text with the
left-mouse button dragging the information to be copied, you can paste
it almost anywhere that accepts text input by moving to the appropriate
text interface, positioning the curser, and prissing the middle button
(both buttons on a two button mouse). This works nearly anywhere that
text input is accepted. Furthermore, you can select text such as
HTML documents on a browser, and get the "simple text" on a simple
text editor.
Some of the interfaces need keyboard shortcuts, but this is as much
a function of application programmers who may just be trying to
"get it out the door quick and dirty" as trying to "get it to be better
than Microsoft's". Most of the commercial software (including much
of the Qt/KDE stuff) is pretty good at this.
> >To say that the GUI designers in Redmond
> > are the only ones qualified to produce an easy-to-use interface
> > is just silly.
>
> Well I am glad that I have never said that as it indeed would be dumb.
Good, we agree on something :-)
> > The best part of all these interfaces is that they are completely
> > interchangable. I can launch KDE applications from Enlightenment,
> > I can launche GNOME applications from KDE, and I can launch Athena
> > applications from Afterstep, and you can easily use any of the above
> > combinations interchangably. This is because the ICCCM formalized
> > the standards of communicating between windows clients. This makes
> > the alteration or substitution of desktops and window managers
> > transparent to both the user and the application.
>
> Yes they are interchangeable, but are they consistent from the point
of
> view of how the user interacts? No, and this may not be a problem, but
> when you use different widget sets, there are fundamental problems
with
> how applications can interact.
Actually, there is quite a bit of consistancy. The ICCCM assures that
buttons do what buttons do, reguardless of where the button lives, or
which widget set is used. The icons may vary, but the core building
blocks are pretty consistent and nearly identical to Windows/Mac/OS2/...
Most of these core building blocks were defined by SmallTalk 20 years
ago.
> > If you look closely, most of the WinTroll attacks have been based
> > on the difficulty of installation. This is where the single
> > installation with two sets of dialogues (one at the very beginning
> > and one after all of the software is installed) with no reboots,
> > gives Linux an advantage over equivalent installations of Microsoft
> > Windows and Multiple Applications that involve numerous reboots and
> > manual configuration using an exclusively interactive GUI panel.
>
> Yes. I don't really care that much about installation. I care about
> users who would like to use Linux with little experience. Most of
these
> people could buy a computer with Linux already installed (just as
> windows). I do care about post configuration such as adding a network
> card or configuring IP addresses or setting up printers and Internet
> connections.
Several cards I have purchased recently actually featured
"Linux Compatibility" right on the box. I plugged them in, ran
Lothar (Mandrake's hardware detection program) and sure enough,
it came up with the right settings in one pass. After two reboots,
My sound card was working again too.
> > A GUI is a great thing when you are doing something only once, and
you
> > need feedback at every step, but when you want to do the same thing
> > many times, or you want to schedule the interactive dialogues such
> > that you only need to have one set of dialogues and complete that
> > phase at a later time, Linux used the approach of creating GUIs
> > that create scripts. This gives the advantages of a GUI, with the
> > stability and reproducability of scripts. This makes it easy
> > to configure one machine - using GUIs, then copy the control files
> > and essentially say "just like that, but change my IP address and
> > hostname".
>
> I fail to see what this has to do with anything (while very true of
> course).
As you point out below:
> > > IMHO:
> > >
> > > "Ease of use" to me means that you can get things done faster and
> > > better.
> >
> > On this I would give you unqualified agreement.
>
> Good :-)
And in some cases Scripts, CLIs, and script generators can provide a
better way of getting things done faster and better.
> > > It means "power" to the user that otherwise would not possess
> > > the knowledge to extract that power.
> >
> > Were moving into murky waters here. Often, users shouldn't have
> > too much power if they lack knowledge.
>
> I like murky waters, it's where though decisions must be made.
>
> >If they start routing a
> > 10.*.*.* address out through your T1, that's a problem. If they
> > decide to make up an IP address because they don't realize that
> > each address must be assigned, or worse, they get the IP address of
> > the mainframe, thinking that should be the IP address they are
supposed
> > to use, you can end up with very unhappy system administrators.
>
> Surely the simple unix permissions should help there when combined
with
> configuration tools? You can't stop people doing silly things. I would
> argue though that you should give them the power to _easily_ configure
> things that they should. Of course in a networked environment these
> things are important but lets concentrate on the home user/business
> user.
And there's the rub. Are we talking a system with *NO* administrator,
a local, in-house administrator, or a remote administrator who handles
the installation and initial configuration then supports via remote
access.
> > > "Ease of use" means that Linux can be accessible to people whom
> > > otherwise would be locked into closed source for-profits-only
> > > solutions.
> >
> > Agreed. I'm still not sure that every one will want Linux. There
> > are those who want "Freeze-dried-instant-microvable success, with
> > a lawsuit guarantee", and think they will get it with Microsoft.
> > Even though the track record over a 5 year period indicates just
> > the opposite.
>
> Yes, but my point is that for Linux to become most successful M$
> "newbies" should be able to jump on-board without drowning in hand
> editing text files with vi. As you point out the attitude of M$ users
to
> support and legal redress is questionable (and laughable).
I've successfully configured an entire Mandrake system without ever
resorting to vi on text files. Linuxconf is very good, as is DrakConf.
I've also done pretty well with YAST - the SuSE version of GUI
configuration.
> > > It also means that the MARKET will take Linux as seriously
> > > as it does with a Microsoft or an Apple.
> >
> > The industry is starting to take Linux VERY seriously. When you
> > have IBM, DELL, HP, Compaq, Sun, Gateway, and Sony, Toshiba, Sega,
> > and Nintendo all considering Linux to be a strategic product,
> > it's foolish not to pay attention.
>
> I am not sure. Show me the drivers..... Until the above companies
start
> putting code where their mouths are then I'll be impressed (call me
> cynical). It seems it's always up to the Linux community to write the
> code - so how far do these companies "commitment" go? Releasing
products
> specs is a vital first step, but it seems that only IBM have made any
> "real" commitment.
Actually, IBM, Dell, and Toshiba have openly contributed code to
Linux for things like drivers. Maddog Hall, a well-known Linux
advocate and "guru" works for Compaq and regularly contributes
(and routes other contributors at Compaq). Michael Dell was a UNIX
fan when he first started the company and has encouraged the use
of "Linux Compatible Components" for at least 3 years that I know of.
Dell employees have often contributed to Linux driver efforts.
The Ipaq hand-held computer also runs Linux.
Many companies have been very covert, not wanting to risk their
relationship with Microsoft. Others have been public, but haven't
been pouring tons of money into advertizing.
At both the PC Expo in NYC and the Linux Expo in San Jose, these
vendors showed Linux machines as key products to be on the shelves
during the "Back to School" period through the Christmas season.
Many vendors were holding off until the 2.4 kernel is officially
released.
> > When you see an industry that triples every 9-12 months, and has
> > already sold nearly 6 million units in the previous year, you pay
> > attention. There's a good chance that the number will be 18 million
> > this year and 54 million next year. That gets noticed quickly.
>
> This will not continue if ease of use is not a mantra for the Linux
> community and that requires a lot of people to re-think their
attitudes.
Agreed. There is a great deal of effort going on in this direction.
OEMs are offering Linux based products, many are also offering Linux
Certified products (known to auto-install correctly with little or no
user intervention). LinuxCare has a directory of Linux certified
products. Some PCs in the near future will have a Linux logo such as
Tux or RedHat on the case next to the "Windows Certified" lable for
Windows.
> > When you have an industry in which freely distributable software
> > has created and fueled growth which indicates that roughly 3
> > installations are made for every copy sold and shipped commercially,
> > and apply that to the numbers above, you suddenly realize that
> > you could have 150 million people using Linux within 24-36 MONTHS.
>
> Maybe, but then we don't know (or should care?) for sure.
Sustaining growth of this sort becomes a combination of user demand
combined with supplier alliances created to meet the demand. If
the supply chain breaks down (Linux suddenly becomes unusable), demand
will either deminish or create new suppliers to meet the demand.
Applix, Corel, and Sun/StarOffice have improved their products
substantially over the last 18 months. Even many of the OpenSource
products have improved radically.
> > Thank GOODNESS you don't have to completely retool everything.
>
> Quite.
>
> > > It means that even more device makers will
> > > positively want their hardware to work to
> > > its fully optimised potential under Linux.
> >
> > I think this is already happening. Furthermore, Linux is
> > creating demand for after-market products such as realmodems,
> > Linux compatible video and sound cards, and even Linux compatible
> > USB peripherals.
>
> What? What? Linux compatible USB devices demand????? More real modems?
> Give me a break. Come on 2.4 is all I will say to this. Many companies
> have seen the benefits but some still hold out (come on ATI!). It is a
> sad fact that more and more companies are choosing to close their
> product specs and move to closed software emulation. IMHO while there
> has been great advances in the broad range of supported devices it
needs
> some real market demand to really start the Linux snow-ball effect.
While many of these companies are tightening up their specifications,
many of them are offering drivers, in either binary or source code
form, which are being included on the major commercial Linux
distributions. The IHVs, especially the after-market IHVs compete
for market share, and Linux compatibility seems to impact that market
share enough that no one wants to cut themselves out of that market.
Many USB peripheral makers are announcing Linux support as well.
There are patches for support of dynamic hub configuration which
will be more fully supported with the 2.4 release. Even the later
2.2 kernels provide reasonable USB support.
> > > It means that software producers MUST consider their
> > > software portability as the market share justifies investment.
> >
> > Fortunately, the Linux/UNIX community has been trying to make
> > this easier instead of harder. Most of the tools used in Linux,
> > such as PERL, Python, Java, and even GCC/EGC are now available
> > for Windows as well.
>
> Agreed, but the user base must be their for the manufacturers to make
> use of this. This requires full desktop user friendly distributions
that
> are not made by Corel.
If using "Linux tools" meant the exclusion of Windows, this would be
true. But most of these tools can be used to create both Windows
products and Linux products from the same source code, often without
modification. Red Hat is offering CygWin, which makes running Linux
applications possible. Other languages and tools are also available
on BOTH platforms.
Ironically, people are finding that it actually costs less to build
for dual-platform than it does to build on Windows-only toolkits.
> > Meanwhile, Microsoft, attempting to continue to hold a captive
market
> > at bay, has been trying to convince ISVs to use even more
proprietary
> > APIs and to become even more dependent on Microsoft.
>
> Yes well we all know that M$ is pure evil.
No. They are a corporation which has a PE of nearly 100 based on it's
ability to grow at 20%/year. For 20 years, there has been no
significant
competitor in their key strategic markets, and they now have a viable
competitor which has actually impacted them enough to reduce their
revenue.
Bill Gates isn't even, but he's not a fool either.
He knows what Linux is, he knows that Linux is a threat,
he was aware of that threat as early as 1993, and has
taken partially effective measures to mitigate the risk
and control the damage. It's what you'd expect a good
CEO to do.
The fact that some of those measures were illegal is something
the courts are handling right now.
> They just make things harder.
> M$: Look at corba that's a neat idea!
To Microsoft, CORBA meant a key infrastructure that could easily
result in the migration from NT server to UNIX servers, and from
NT workstations to NCs such as JavaStations or LinuxStations.
Just as the early web browsers moved much of the developement effort
from Windows to UNIX, Corba could have resulted in increased mindshare
for Linux/UNIX.
> Bill: Yes, lets define our own
Bill needed something that could support UNIX, but wouldn't
turn the control of the development cycle over to UNIX. In
the Netscape/Visigenics model, the browser was an extension of
CORBA which primarily supported UNIX servers.
Gates responded with the DCOM/ActiveX model which attempted to
keep control under the domain of Microsoft.
> M$: bring it on master.... It shall be done. We are one.
I've never heard Microsoft employees call him "Master" or "My Lord",
or "your highness". They usually referr to him as "Bill", as in
"O.K. Bill, whatever you want".
When Bill owns 25% of the company, can't be fired, and has complete
control over not only your career at Microsoft, but also over
intellectual property rights to software you create after leaving
Microsoft (part of the employment contract), you have to think very
carefully before telling him "No Bill, I won't do that".
> > > It means the GNU-Linux gets better. Better is good for us all.
> >
> > GNU-Linux is getting better. One of Microsoft's biggest problems
> > is that Linux is such a moving target. By the time Windows 2000
> > came out, Linux had already eliminated most of the performance
> > bottlenecks in the Linux 2.2.11 kernel. The 2.4 kernel is due
> > out in a few weeks and has been delivering performance numbers
> > that are staggering. And 64 bit Linux is already a reality for
> > SPArC, Alpha, and PPC, and just awaiting Silicon from AMD and INTEL.
> > In fact, the 64 bit market could heat up very quickly if Linux
> > starts creating markets for Non-Intel platforms such as Transmeta,
> > ARM, and MIPs chips. Low-cost 64 bit chips could have a huge impact
> > on the "WinTel" monopoly.
>
> Yes Linux is a moving target and "behind" others, but that's my
problem.
Behind or ahead? Many so-called Microsoft "innovations" existed in
a more general form under Linux and/or UNIX as much as 3 years before
Microsoft announced their version (complete with it's own proprietary
acronyms, terminology, and calling conventions).
> In some cases Lin-advocates claim that M$ does not matter and in the
> next breath claim that it must compete. What I'd like to see is Linux
> (Linux+GUI+config tools) getting to the stage where they are reliable
> and easy to use while not taking the raw power away from us who like
to
> bugger about with textfiles and to understand. I don't care what
Windows
> does, just that Linux is "newbie friendly" and that is a term that
will
> have a lot of people here upset because it means taking some pragmatic
> decisions about how things work.
I think the critical point will be the delivery of preinstalled Linux
software in OEM produced/configured machines. If you've ever tried
to install Windows NT in custom-configured hardware, you'd know that
Linux has many advantages over NT, especially in regard to
"post-install"
upgrades. One of the most unpleasant experiences was trying to install
NT on a VLB 80486DX/100 machine. The thing would just BSOD with nothing
running but explorer. It wasn't even connected to LAN/WAN.
> > But Linux keeps evolving in the GUI and Ease of Use arena too.
> > KDE is now offering features that aren't even available on Windows,
> > and GNOME has created software that implements technology that is
> > still on Microsoft's "drawing board".
> >1Microsoft talks about things
> > like Application servers while Linux provides the ability to run
> > nearly any application from either workstation or server.
> > Furthermore, KDE and GNOME have shown that the CORBA model can be
> > very powerful in it's own right. Not only has this enabled the
> > development of remotely accessible applications, but it has also
> > enabled single applications to benefit from the power of large
> > clusters of servers.
>
> yes all true. How about consistency? Gnome is trying to address this
> sure, and I'd love for people to remember the non technical user (why
> was Eazel formed?).
There are some things that are done differently under X11/GNOME/KDE
than under other platforms. For example, X11 provides it's own
clip-board
which can handle many Cut/Paste features. In some cases, you need to
know which form of object you are cutting and pasting.
For example, if you mark a web page and "copy" it, and then you try
to paste it to the text file, do you want just the ASCII text without
any markup, do you want the HTML markup, or do you want the bitmap
for the page in uuencode format.
One of the key differences (and key advantagees) of Linux/UNIX is that
you have enough knowledge of the data to be able to perform transforms
on the data. For example, a set of simple filter programs can convert
bitmaps from bitmap (BMP) to pixmap (PIX) to GIF, to JPEG, to EPS,
and back. It's effectively a crossbar switch.
So the question becomes "in what formats is this object currently
available", and "in what formats could it be delivered to the
destination.
Going back to the HTML document again. If I just want to copy the link
to the GIF file from one HTML document (in a web browser) to another
HTML document (in a text editor), I might just want to copy the
<a image=...> line. On the other hand, if I were using a word processor
to create an RTF document, I might want to embed the actual binary
object within the document as an embedded object (microsoft style).
This is extra complexity, but it is very similar to the issue of
cut/pasting numeric values from a word processor table to an excel
spreadsheet. Typically, you have to select the table, mark it,
paste it into the spreadsheet, and then do "convert text" to fan
it into the cells.
In UNIX and Linux, such transforms are quite common. The job could be
done with a script, or it could be done with a gui that manages the
cut-paste, and furthermore, the object itself may need to offer
it's different copy-paste options.
Now, that was simple copy-paste. With drag&Drop, you have to come up
with a matrix of which input formats are available, and, for any
selected
input format, what are the available outputs.
Windows could get away with trivial single meaning objects because
it was created for single system environments where only one object
class could create, modify, or even import/export that data. With
UNIX, each drag/drop cut/paste is essentially two import/exports.
One could reasonably argue "is this a good thing?", the frame-maker
school says create a frame, put the object in the frame, and let
the object provide the methods. Unfortunately, this means that each
user must have the same library on their machine (part of the reason
for the relatively small number of object types in standard HTML v3.
HTML-4 went plug-in happy and those plug-ins have been a problem.
The good news is that Linux does provide an immense object library
in it's "standard objects". Between KDE, GNOME, Swing, and some of
the older libraries, you can do pretty much anything using standard
object formats. Again, the tricky part is getting the output of
object A to the input of object B when A and B are two different
classes.
This is by no means a new problem. Mac suffered with the opposite
problem. For example, there was no way to get cells from McProject
into MultiPlan. This meant that even if you wanted to create your
own formatted summary report for the project, you couldn't.
> > > Surely the intelligence of any system, whether it is a fly-by-wire
> > > aircraft control system, a video camcorder or a computer system is
in
> > > HIDING the complexity of the tasks that they perform? And surely
the
> > > beauty of Linux is to EXPOSE complexities to those of us who enjoy
> > > them?
> >
> > This is true. Even in a device as complex as an automobile, there
> > are thousands of complex parts engaging in millions of complex
> > interactions per second. To the user, he still experiences the
> > same driving controls he had on his dad's 53 Chevy. But to the
> > mechanic, everything has changed radically under the hood. The
> > "tune-up" is done every 2 rotations of the engine, by a computer
> > that needs to be reprogrammed based on the engine wear. The
> > carburetor has been replaced by a fuel injector which supplies
> > microscopic amounts of fuel to each cylinder in a place and at a
> > time that will assure the maximum vaporization and the most
effecient
> > burn. The two valves per head have been replaced by 4, and the
> > camshafts have also increased from 1 to as many as 4.
> >
> > On some cars, you can't even change the spark plugs without
completely
> > removing the engine. On others, you can replace even the timing
> > belts with almost no special tools.
> >
> > The result, to that user, who still thinks he's driving the same
> > kind of car that is grandfather drove, is an automobile that gets
> > 5 times more distance per gallon of gas, emits 3 times less air
> > pollution, and runs over 120,000 miles without a significant repair
> > bill. Unlike Grampa's Chevy, today's driver is very unlikely to
> > need more than oil changes, tire changes, brake linings, and some
> > occaisional check-ups and inspections.
>
> OK, this is a good analogy. I would like to see learner drivers
quickly
> learning how to use an easy interface and simple configuration (such
as
> topping up the oil and adding anti-freeze easily).
Good. I'd just like to START by formalizing the transition from
"standard" to "automatic" ie Windows to Linux. Some of the simple
"gotchas" like double-clicking, and double-clicking again because
you didn't immediately get a confirmation window is a classic.
By letting the user understand that the reason they didn't get a
confirmation was so that they could get back to whatever they were
doing before (or some other useful task) until everything is
initialized.
Learning a new habit takes 14 repititions. At one time, people faulted
netscape and mosaic because it started loading the link after the first
click. Today, we don't even think about double clicking the links on
the browser.
Unlearning an old habit takes about 30 repititions. This may be
neccessary for things like "chipmonking" (the way a chipmunk fills
his cheeks with nuts so he can get more nuts while supply is plentiful).
This is my term for the practice of doing something useful while your
new program gets initialized.
Ironically, it takes almost the same amount of time to launch a program
on Windows as it does on Linux (for the same machine). Microsoft gives
you passive entertainment, UNIX lets you get back to work.
Rex Ballard - I/T Architect, MIS Director
Linux Advocate, Internet Pioneer
http://www.open4success.com
Linux - 42 million satisfied users worldwide
and growing at over 5%/month! (recalibrated 8/2/00)
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
