1Welcome to the OS/2 Warp Monthly Newsletter

The theme of this month's newsletter is Leaps of Technology.  In case 
it isn't obvious why Ichose this as the theme for the February issue,
I'll explain.  This year, 1996, is a leap year.  I don't mean to insult
you by informing you of this but when I told people about the theme for
this issue, they were surprised that this year is a leap year.  I think
the fact that we now have an Olympics every two years may be confusing
everyone.  Ormaybe it's just that people don't realize the year is 1996
in the first weeks of January.  Or maybe I just talk to some pretty
unaware people.

Even so , the February issue is the "leap" issue because February is
the "leap" month.  Actually, never in my life, before now, had I
referred to February as the "leap" month.  But when I said this to
people, they nodded their heads like they had always thought of
February that way.  (Then again, maybe they wanted to recover from
being unaware that '96 was a leap year.)

Speaking of February, I never have understood why January and March
both have 31 days and February, on a regular year, only gets 28.  If
both January and March each donated a day to February, then all three
would have 30 days.  February would not need the extra day during leap
years and January, the first month of the year, could get the extra
day.  Doesn't that make more sense?

So far this discourse has not had a lot to do with technology or OS/2
Warp but I'm getting there.  However, it's going to take a few more
sentences, so bear with me.

Now that I've established the validity of the "leap" theme, let me
explain what I think of when I think about leaps.  I think about
ballet.  To understand why I make this connection, check out a picture
of John Renvall performing a leap in the role of the Bronze Idol in La
Babyadere at
http://www.i3tele.com/photo_perspectives_museums/faces/labaydere.html
or other leaping American Ballet Theater dancers at
http://www.i3tele.com/photo_perspectives_museums/faces/butler.html.

Dancers, trying to perform a perfect leap, want to create a sense of
ballant.  The literal translation of the French, ballant, is dangling,
but dancers take it to the next level.  To them, ballant is the ability
to appear as if they are effortlessly suspended in the air and could
remain there forever.  This is as opposed to what they are actually
doing; using a great deal of strength, balance, and determination to
fight the forces of gravity.

It seems that we can apply this same term to leaps of technology.  The
innovations or products that embody a successful leap of technology
will also create this sense of ballant.  That is, they would be so
obviously the next step that everyone thinks, "Why didn't I think of
that?".  Almost as if the idea was dangling in the air and anyone could
have grabbed it.  Furthermore the products with ballant would have
longevity at the peak of the technological leap both in the original
form and in the ability to expand and grow the original concept.

OS/2 has the characteristic of technological ballant.  It was most
definitely the next obvious step after DOS -- it corrected many of the
DOS shortcomings.  It has also shown longevity.  The first version of
OS/2 was released in 1987 and, contrary to what some of OS/2's
opposing forces would like people to believe, it is still going strong.
Furthermore, it is continually being extended.  Not only does IBM
extend the original concept with new features and functions but
application developers expand upon it using OS/2 technologies to create
a wide range of innovative products.

I know you're thinking that after this introduction, this issue better
be good.  Don't worry, it is.  We look at the technological leaps
included in new OS/2 products including OS/2 Warp Server, Lotus'
SmartSuite 96 OS/2 Warp Edition, and Notes Release 4.0.
We even have an article on the Single Input Queue (SIQ) fix that we'll be 
shipping in the next release of OS/2 Warp written by developer, Aidon 
Jennery. We also have some good stuff in the Shareware and Commercial 
Applications section including a description of Athena Design's Mesa
Developer Object Kit.  The other sections are meaty as well with tips 
on correcting LAN Server errors, the second of Jose deJesus' Java
articles and an indepth interview about OS/2 with computer crime
researcher, Professor Nikolai Bezroukov.

SHAREWARE AND COMMERCIAL SOFTWARE
 From object technology leaps, AthenaDesign bounds 
 by Jeri Dube

From a software perspective, one of the greater leaps of technology has
been objects.  While they have been around for years, only now are we
seeing the technological leap gain altitude as more developers
incorporate object technology into products.

Athena Design, best known for its Mesa spreadsheets for OS/2 Warp, has
always used object technology.  Now Athena is going a step further and
providing its customers with a Developer Object Kit.  The Mesa
Developer Object Kit (DOK) contains the same 32- bit object components
that Athena used to build Mesa.  It also includes the file filters that
allow read and write compatibility with other spreadsheet applications
such as Lotus 1-2-3.  The Mesa DOK now allows application developers to
add spreadsheet fun ctionality to other applications without
programming the details themselves.

Already Mesa DOK beta customers are adding spreadsheet functions to
medical systems, real-time data feeds and analysis systems and a range
of applications important to sectors of the finance industry.  To add a
spreadsheet to a C++ application, you instan tiate a model (from the
spreadsheet engine class) and then connect this model object to a view
object.  The view object displays the spreadsheet in a window on the
screen.  This example requires only about 12 lines of code.  VisualAge
C++ users can create th e same functionality using drag and drop.  All
the user would have to do is drag a spreadsheet component from the Mesa
DOK and drop it on the project.

The object technology allows Mesa DOK users to customize the
spreadsheet components to accommodate individual requirements.  For
example, if the user had unusual display requirements so he could
easily compare spreadsheets, he could make modifications with out
affecting how the rest of the functions work.  The user would also
minimize his programming effort by creating subclasses that inherit
many usable attributes and methods from a parent class.  The user would
then modify only some of the methods associated with the new
subclasses.

The Mesa DOK is based on IBM's System Object Model (SOM).  This makes
it easy to distribute Mesa DOK upgrades.  When a customer gets an
upgrade, he does not have to recompile nor relink the applications that
depend on the Mesa DOK.  Future versions of Mesa DOK will be
backward-compatible.
 
This spreadsheet object kit is the first of its kind available to OS/2
users and developers.  You can grab a free beta copy at
ftp://ftp.athena.com/pub/Mesa2/dokpr1.zip.  Athena and resellers will
begin selling the first release some time during the first quarter of
1996.

Product Review:  Technical Connection Personal Software


IBM Technical Connection Personal Software is a collection of tools
that allow users to have quick access to the latest in service and
product support information.  And what a truly wonderful collection it
is.

The software collection consists of seven components including AskPSP,
Solutions Database, Technical Bulletins, Corrective Service Diskettes,
Reference and Diagnostic Diskettes, Product Information, and Device
Drivers.

The Technical Connection is easy to load and even easier to use.  To
install it, you simply put the CD-ROM into the system, change to the
drive letter designation for your CD-ROM and type install.  It really
couldn't be an simpler.  The icons in the package are well labeled and
easy to understand.

If you own IBM equipment, the Reference and Diagnostic Diskettes are
indispensable options.  It could turn out to be very handy the first
time you can't find your reference diskette for your PS/2 Model 70.
The list of equipment diskettes is extensive enoug h to include CD-ROMs
and page scanners.  I was quite impressed.

The Product Information section will keep you up to date on the latest
IBM products available.  It also includes IBM Dealer Support News, Team
OS/2 Newsletters, and several other newsletters from different groups
at IBM, including this one.  Demonstration software programs are also
available here from time to time.  This area contains such a tremendous
amount of information that it alone makes the Technical Connection
subscription worth the price.

The Device Driver section is also indispensable for medium and large
companies that use a wide variety of equipment.  The video driver list
contains drivers from ATI, Western Digital and Cirrus Logic.  The
PCMCIA section contains drivers from several manufa cturers including
Toshiba and Apex.  Similarly, the CD-ROM section presents drivers from
Mitsumi, Sony, Phillips, Chinon and others.  As a technician, this
section really caught my attention because it would save me from having
to call a half-dozen differe nt bulletin boards to get the drivers I
need.  It's a real time saver.

The Technical Bulletins section is divided into PS/2 Assist and OS/2
Assist.  The PS/2 area contains the PS/2 Assistant, the Consumer PC
Assistant, PS/2 Marketing, and General Information Assistant.  The OS/2
Assist area is an exceptionally comprehensive listing that includes
the OS/2 platform product strategy, questions and answers on several
software products, employee written software (that does a wide variety
of things), the Personal Systems support family (PSSF) problem
reporting guide and users guide , and last, but not least, the
Technical Solutions Library.  The Tech Solutions Library contains
information on OS/2 Warp, Communications Manager/2, DOS, LAN, and
multimedia.  All in all, it is a very complete guide that complements
the other sections very well and provides yet another source of
solutions.

To stay current with the latest fixpacks and service packs, jump into
the Corrective Service Diskettes section.  The Technical Connection
compresses over 350 megabytes of information and fixes into this area.
Most of the fixes concern installation and crea tion instructions.  The
fixes cover a wide variety of products including Personal
Communications AS/400, OS/2 Warp Connect, and Communications Manager/2.
This is another incredible grouping of information and tools to make
your support technician's life a lot easier and keep your systems
running smoothly.

The Solutions Database allows you to find solutions to reported
problems.  This is the area where you can get confirmation that you are
not the only person in the universe with this problem.  This database
is a collection of APARs (Authorized Program Analys is Report).  Each
entry contains a brief description of the closed problem, the symptoms,
and the resolution.  Previously you could only access this information
on a BBS, now you can have it at your disposal at any time.  This
definitely provides a savings i n time and money as well as increased
efficiency in getting a problem solved without having to call for help.

I have saved my favorite part of the Technical Connection for last.
AskPSP is your own personal help desk.  It uses an expert system tool,
CasePoint, to allow you to ask questions in everyday English to help
you find solutions.  It contains case bases on a variety of products
including Communications Manager/2, OS/2 Warp, OS/2.X, DCE, LAN
Systems, OS/2 Warp Connect, and	 Novell NetWare from IBM.  You will
also find a readme file and a tutorial to show you how to use AskPSP.

For example, imagine you were using OS/2 Warp Connect and could not get
the software to see your 3COM 3C509b network adapter.  By describing
the problem to AskPSP and answering its questions, in a matter of
minutes you would find that the card probably is not configured
correctly.  It would tell you to check the transceiver type (set to
match the cabling that you are using), to disable the plug & play
feature, and to set the modem speed in the softset configuration to
9600 baud.  It is really a phenomenal to ol that many of the support
technicians at IBM actually use.  I require it in my bag of tools.  A
lot of hard work has gone into creating the case bases and we are
continually updating them.  The effort and commitment to quality are
evident.

The cost of the Technical Connection is quite reasonable.  A single
purchase costs $59.  An annual subscription is available with the
updates coming either monthly or quarterly.  The quarterly updates
subscription is $119 per year while the monthly updates subscription is
$299.  As an initial purchase bonus, the Technical Connection includes
a copy of the CD-ROM containing over 275 books on a variety of OS/2
products.

Having been a support technician with several hundred systems to
maintain, I wish I had had a tool the caliber of the Technical
Connection.  I have not seen a support tool of this quality from any
other hardware or operating system manufacturer.  I highly recommend
it to both end users and in-house technicians alike.

 

* This article is based on the author's personal review of the product
and should not be viewed as an IBM endorsement of such product.

Product Review:  Xit 2.0 by Collin Summers
 
One of the neatest little shareware OS/2 Warp utilities I have seen in
a while is Xit.  It is the follow-on to the freeware Xit that added a
one click close button to your title bars.  Xit 2.0 goes beyond the
close button and allows for the addition of all kinds of different push
buttons to title bars as well as adding a few other functions.

Let's start off by going over the different function buttons that you
can add.  Lots of buttons allow easy one click access to different
options.  Without Xit, many of these options are hidden in the top left
icon's pull down menu as well as in other pull-down or pop-up menus.

Here are my favorite functions.	

The roll up/down window is a great option that literally rolls the
window into the title bar.  It's great for cleaning up the screen.  The
user command button is one of the most useful options.  The user
command option allows Xit to run anything you can currently run from
the command line.  I have it run my personal information manager; so
its always easy to launch no matter what I am d oing.

Some of the other one click button options are:  The close window that
lets you close windows.  The Xit task list pops up an abbreviated
version of the OS/2 Warp task manager -- or you can use it to bring up
the OS/2 Warp version. Extended, keys, or index help buttons all
allow for one click access to different sections of the help screens.
The next or previous window lets you cycle through open objects.  The
move or size window lets you quickly invoke the moving or resizing of
windows.  The show launch pad will bring the OS/ 2 Warp toolbar to the
top of your desktop.  The edit popup menu brings up some of the typical
edit commands, such as cut, copy, and 	 paste.  The mark and copy to
clip board, copy all to clip board, or paste from clipboard do just
what they say they'll do with one click.

For Presentation Manager windows, you can add buttons to clear/delete
selected text or cut to clipboard.  In DOS and OS/2 windowed sessions,
you can open the set font size dialog box.  You can also put a DOS
window into full screen mode.  	 Xit makes it a lot easier to
manipulate files and folders.  You can do many things such as navigate
between folders with the open parent or open parent and close current
folder buttons or arrange and refresh now the objects in the folder.
Xit buttons also make it easier to get to sort by:  name, type, real
name, size, write date, access date, or creation date functions.

You can have up to ten of these optional buttons in operation at a
time.  You can set up each button either on the left or the right hand
side of the title bar.

If this sounds good to you , "Wait, there's more."  You can enable
mouse options so that your center or right mouse button perform various
functions.  You can also set the mouse to do any of the functions, I
described earlier by simply clicking the designa ted mouse button.
Similarly, you can make use of the Alt, Ctrl and Shift keys in the same
way when they are held in conjunction with the selected mouse button or
in various combinations with each other.  For example, you could set up
a key/mouse button co mbination to perform the roll up/down window
function.  You can also use Xit to set up your keyboard to act as
double or single clicks on the mouse buttons.

The one downside is that some programs don't seem to work well with
Xit.  That isn't a problem, because an Xit option disables it from
working within those programs.  You can also use this option to add any
applications where you don't want Xit to be acti ve in the title bar.

I am sure I have left out some features but I think you can see it is
loaded with lots of function.  After using it for the past two weeks or
so I have found this to be one of my most treasured add-ons.  (Michael
and Peter my check will be sent post-haste
.)  You can try Xit 2.0 out yourself at ftp://hobbes.nmsu.edu/os2/wps. 



* This article is based on the author's personal review of the product
and should not be viewed as an IBM endorsement of such product.


PRODUCT TIPS Quick Tips by Collin Summers

Work Area Folders:  By using work area folders you can easily launch
groups of applications.  Simply create a folder on your desktop (Drag
one from the templates folder).  Open the folders settings and go to
the file tab.  Click the work folder check box.  Now whenever you open
that folder its objects immediately go to the state they were in the
last time you closed it.  For example, when I am out of the office, I
open my Mobile Office folder.  It launches Passport/2 (to connect to my
host system), opens my document folder, and launches E-mail
applications.  When I close the folder, the command closes all the
objects.

Multiple Instances of Applications:  You can run multiple instances of
an application at the same time.  For example, you can launch multiple
WebExplorers.  To set this up, go to the WebExplorers settings
notebook.  Go to the window tab and change it from display existing
object to create new window.  Now every time you double click on the
WebExplorer, it will launch another one.

Templates:  You can turn often-used documents into easy to reuse
templates.  Simply create a document or shell of a document using your
favorite word processor (It will work with almost any kind of data).
Go to its settings and select the general tab.  Select template.  You
can now move the template (by using pickup) to an easily accessible
place.  Now, whenever you want to create a new version of that document
just tear it off the template, change its name, and off you go.

Reduced Popup choices:  Hitting the right mouse button gives you a
myriad of choices.  This choice set can be reduced by adding SET
MENUSTYLE=SHORT to your CONFIG.SYS file.

 Correcting Networking Errors for OS/2 Warp Connect by John Crane
There are several errors that may occur on an OS/2 Warp Connect system
attached to a network.  Most of these manifest themselves at boot,
while others hide themselves and you have to search them out.  All
these errors generally fall into one of three categories, LAN
Transport (LT) errors, Network errors, or Requester errors.  This
article covers only the most frequently occurring LT errors.

Since many of these errors may occur quickly during boot, the user
needs a way to be able to find what errors are occurring and in what
order.  OS/2 Warp Connect keeps two error logs, the Net Error log,
which tracks every single error, and the LANTRAN.LOG that only keeps
track of Network Interface Card (NIC) or network connection errors.
You can review the Net Error log by simply entering "Net Error" at a
prompt.  The LANTRAN.LOG resides in the IBMCOM directory, and therefore
is only present when Multi-P rotocol Transport Services (MPTS) is
running.  You can read it by changing to the IBMCOM directory from a
[C:] prompt and using the type command to view it.  You can find any
errors that occur in OS/2 Warp Connect in one of these two logs.

LT 00014 is a seldom seen error that indicates there are not enough
resources available for the PROTOCOL.INI to execute all of the drivers.
In most cases, the most simple solution for this problem is to add more
memory.  If the system already has 32 MB or more, the next step would
be to limit the number of drivers loading in the CONFIG.SYS.

LT 00042 is a common error especially if the OS/2 Warp Connect
installation was not able to "sniff" the NIC adapter.  Usually, this
error indicates OS/2 Warp Connect's inability to bind to the card
drivers or the drivers' inability to initialize the card.  The binding
or initializing process breaks down when the drivers are not written
for the OS/2 Warp operating system.  Most recently manufactured cards
come with OS/2 Warp drivers on a diskette or the OS/2 drivers may be
available from the card manufactu rer's Bulletin Board Service.  OS/2
Warp Connect requires two drivers.  Both have the same name but their
extensions should be .NIF and .OS2.  For instance, the user may have an
IBM token ring card.  The drivers for OS/2 Warp Connect that come with
the card are called IBMTOK.NIF and IBMTOK.OS2.

The LT 80204 error usually indicates that NetWare Requester is trying
to start in MPTS, but the Novell NetWare Client has not yet been
installed or started.  Specifically, it is looking for the LSL.SYS to
have already been loaded in the CONFIG.SYS.  Usua lly, installing the
NetWare Client, by typing Prodinst (Product Install), from the CD-ROM
drive will correct this problem.  If not, check CONFIG.SYS to see where
the LSL.SYS is loading and be sure it occurs before MPTS initializes.
LT 80204 is an error that usually leads to a PRO 0033 error as well.
The PRO 0033 occurs because one or more protocols have not been able to
use the card, as a result of incorrectly loaded card drivers or because
OS/2 Warp Connect could not initialize the card.  This is directly
related to the LT 80204 error and will usually disappear when you
eliminate the LT error.

The LT 80210 error occurs when the physical address for the network
adapter card driver has not been configured for ODI2NDI.  You can find
this node address in the LANTRAN.LOG.  If it's not there, look on the
card itself or run the diagnostic provided with the card to find it.
After you find the 12-digit node address, open MPTS and add the address
to the first line of the NetWare Requester protocol.  If you are
loading 802.2 as a protocol, check to see if it has a different address
configured.  If so, eit her change it to the one found by the above
steps or configure it for the adapter's universal address.  You can
find the universal address only by running diagnostics on the card.
When you exit MPTS, it gives you the option to update the CONFIG.SYS
file w ith the changes you just made.  Save and update the CONFIG.SYS
file and everything should be fine.

LT 80213 is another instance of the NetWare Requester not being able to
bind to either the card or its drivers.  This is a very similar problem
to an LT 80204 and you can resolve it in much the same way.  It is not
unusual for this error to be followed by PRO 0025, PRO 0026, PRO 0027,
or PRO 0029 errors.  These errors involve the various protocols that
are trying to attach to the adapter card, and will almost always clear
up once you resolve the LT 80213 problem.  Remember, there is the
possibility that the NIC could actually be defective.

LT 80215 has to do with the Query Time-out value established by
NetBIOS.  You can set this value in MPTS by editing the NetBIOS
protocol or edit the NETBIOSTIMEOUT variable in the PROTOCOL.INI file.
The values can be set from 500 to 10000 where the default is 2000.
Usually increasing the value over 2000 clears up the problem.

LT 80225 is an error dealing with Frame Types set in the NetWare
Requester of MPTS.  The default Frame Header Type is Token Ring, which
if you are using an Ethernet card will cause this error.  If the
network is using 802.2 or 802.3, you can set that here as well.  Once
again though, if you set it incorrectly, it will cause problems.  You
can only set one of these protocols at a time, as this is a selective
choice.

LT 80227 is a similar error to LT 80210 and you should follow the same
steps to resolve the problem.  This error deals only with the NetWare
Requester so how the 802.2 protocol is set does not affect it.  The fix
does not require an universal node address. Just get the adapter node 
address from the LANTRAN.LOG and add it to the NetWare Requester 
protocol.   

You can resolve these common LT errors using the methods described in
this article, although you may encounter others.  Do not forget that
help for any error is available at a command prompt.  Just type help
and the error.  For example, to find out more about an LT 80225 error,
at the [C:] prompt type help lt80225.  Notice there is no space between
the LT and the number.  These tips should get you through most of the
LT errors that you can encounter in OS/2 Warp Connect.  Laptop
deployment in the workplace by Kurt Ruby and Diane McBride

There is little debate that laptop computers can increase productivity.
Our jobs within the IBM corporation provided us the opportunity to
deploy laptop technology in the workplace.  We first deployed a pool of
IBM ThinkPads for use on business trips.  This later led to deploying
IBM ThinkPads as primary workstations for select people in that same
development organization.  We learned a number of lessons as we rolled
out these two usage types.  It is our hope that other people attempting
a similar imple mentation can learn from our experiences.

Hardware Configuration Before diving into the lessons, a little
information about the hardware and software configurations we used
might provide some insight.  The travel pool of laptops consisted of a
hodge-podge of various IBM ThinkPad models (i.e.  models 350, 500,
360C, 360C E, 755C, 755CS and 755CV).  The memory capacity ranged from
as little as 8MB on a ThinkPad 360 to as much as 40MB on a ThinkPad
755CV (this model has an overhead display).  The hard disk capacity was
either 340MB or 540MB.  All machines had PCMCIA modems and a few had
PCMCIA token ring cards.

The laptops deployed for workstation use were ThinkPad model 755's.
They all had 20MB RAM, 540MB hard drives, PCMCIA modem, PCMCIA token
ring cards.  These workstations included IBM Dock II stations that
provided speakers, two ISA slots, two additional PC MCIA slots, CD-ROM
bay and hard drive bay.

Software Configuration All the laptops had OS/2 Warp FullPack (with
IBM's WINOS2 code) installed along with various software products.  Our
users needed to run OS/2 and a few Windows 3.1 applications.  This made
OS/2 Warp FullPack a natural choice.  We chose the FullPack version
over OS/2 Warp because we would only have to install a single operating
system.  Although most of our laptops came preloaded with Windows 3.1
and/or OS/2 Warp, we chose to reformat the hard drive.  We felt that
because the majority of pre-installed sof tware was targeted for home
use it was not much use in the workplace.  We were rewarded with a
smaller hard disk footprint that gave the user more space for data and
other applications.

As an added bonus, the plug-and-play support that OS/2 Warp provides
for PCMCIA cards and other ThinkPad hardware features impressed us.
(This was not the case with Windows 95).  The support made software
configuration fairly straightforward.

Common Lessons As we look at the two types of laptops deployed, some
common lessons apply to both types.

Limit the number of laptop models if possible.  We probably discovered
this lesson first.  Some laptop models of ThinkPads (and probably other
non-IBM laptops as well) have a sl ightly different hardware BIOS.
Differences in display technology also made software compatibility
impossible.  The impact was having to learn and know about all the
hardware variations therefore creating higher technical support costs.
  
This lesson is not always possible to follow due to budget constraints
or simply models being obsolete.  The trick is to minimize the
variability within the hardware.  We also found that having less
machine models will reduce the cost of disaster recovery (not in the
classical sense).

Install anti-virus protection software.  Anti-virus protection is an
absolute must for laptops, traditional desktop PCs or workstations.
The Internet and other on-line services make it so easy to introduce
infected software on a PC that many companies make this a requirement
on all machines they own.  We use IBM Anti-Virus for OS/2 and update
the software as recommended.  (It actually tells us when it needs
updating).
   
Partition the hard drive as a single primary partition.  How you
partition depends on the capacity of the hard drive.  Our experience
was that more than one partition on the disk caused lots of technical
support questions.  People not knowing or realizing that there was more
than onepartition caused many problems. Both the travel pool and
workstation users would install additional applications, quickly
filling the C drive (usually 100MB or so).  OS/2 Warp would then issue
warning messages about not being able to save the INI files.  Using a
single partition solved this problem.

Have a simple disaster recovery strategy.  This is one of the oldest
lessons taught and is also the most obvious.  The advantage of having a
simple disaster recovery process is twofold.  First, the disaster
recovery process makes it easier to reload the "travel pool" machines
when people return th em.  We learned that most people tailor the OS/2
Desktop and load some custom applications after they check out the
machines.  When users returned the laptops, it was unlikely that the
original desktop or application base was still on the system.  A 
simple disaster recovery plan in this case made it easy to reformat 
and restore the original software image from a tape backup.  (Also the
fewer model variations, the less number of tape backups -- see the
first lesson).

The second reason for having a simple disaster recovery strategy is for
the laptop workstation user community.  These people can have a backup
available in case of a hardware failure.  We use the IBM ADSM/2
application to incrementally do backups using a token ring LAN
solution.  All laptop workstation systems also have software installed
for a parallel port backup tape unit.

Provide security devices to protect against theft.  Since laptop
computers are mobile, they are the target of thieves.  A number of
products exist on the market that provide reasonable security devices
for various makes of laptop computers.  Some models of the IBM ThinkPad
provide security enabling devices that attach directly.  We do not
recommend any specific product, but keep in mind these devices do not
guarantee 100% that they will deter thieves.  Based on the products we
have encountered, a determined thief can still defeat these products.

It is important to note that the IBM ThinkPad 360, 755 and later models
have the system serial number as part of the BIOS information.  This
can be an aid when trying to retrieve a stolen machine.  Alerting IBM
service centers of the serial number could result in recovering a
machine, if and when it is brought to a service center.

Use power-on passwords or additional security features if the laptop
provides them.  Some laptop machines provide some level of hardware
power-on password to protect data and/or hardware settings from
unauthorized use.  Power-on passwords can usually be removed via jumper
settings or removing the battery for extended periods of time.  Wh ile
not fool-proof, it can be a significant deterrent to theft.  Some
laptop models provide greater levels of hardware password protection.
For example, the IBM ThinkPad 755 series supports a power-on password,
a hard drive password and a system password.  The power-on password 
can be disabled by IBM service, while the hard drive and system 
password cannot be overridden.  The latter two password schemes 
provide a high level of security, but they requiring careful planning 
before use.

Establish a stockpile of hardware/parts that commonly fail.  In the
perfect world, nothing ever breaks.  Hard drives, diskettes, and PCMCIA
cards should theoretically work forever.  This is sometimes the case
with desktop PCs but laptop computers were meant to be portable, which
implies that they get moved around, sometimes dropped or, even worse,
crushed.  We found that simple things such as PCMCIA modem cables break
with normal use.  Below is a list of some of the items we stockpile to
reduce downtime:  modem cables, splices, extension cables power
supplies and cables batteries hard disk drives token ring (LAN) cables
diskette drives

The quantity to stockpile depends on the number of laptops being
supported.  People using laptops as primary workstations in the
workplace sometimes like an extra power supply for work only.  (This
way, they don't have to handle a power supply at home and at work).

Have adequate technical support.  Laptop computers, while similar to
their desktop brethren, require special technical support.  PCMCIA
cards provide challenges both from a hardware and software standpoint.
The BIOS is usually different than those found on desktop systems.
Loaning new l aptop machines to the technical support staff when they
arrive helps them develop the expertise they need.

Travel Lessons The "travel" pool of laptops provided us our first look
at laptop technology in the workplace.  Everyone who used our pool of
machines found them to be a big productivity aid when traveling on
business.  This deployment taught us a number of lessons uniqu e to the
mobile user.

Have a well defined check-out system.  A check-out process is one of
the obvious necessities when loaning machines.  Scheduling machine
availability is not usually sufficient for tracking purposes.  We found
that a property pass signed by the people checking out machines made
them responsible for the asset and provided proof to site security that
the equipment could leave the premises.

In addition, we designed a custom Lotus Notes database to track the
hardware.  It provides a means for scheduling, describes each machine's
hardware/software characteristics, and a place to store the current
power-on password (new passwords are assigned w hen machines are
returned).

Have a fast and reliable means to reload laptop software.  The simple
backup/disaster recovery process discussed in the previous section
served this purpose.  Since people rarely return a loaner machine with
the same software or the same configuration, we reformat the hard disk
and reload the base image from tape backup, just as we would do in a
disaster recovery situation.

Install common and popular applications.  We discovered quite early
that people needed a core set of common applications.  For our
purposes, we provided the Lotus SmartSuite for OS/2 (and Windows
SmartSuite on some ThinkPads) along with a remote connectivity
application (PASSPORT/2) to connect to the IBM network . We suggest
loading the enterprise-recommended word processor, spreadsheet and
connectivity software on all travel systems.

Provide a good quality carrying case.  This may be one of those
overlooked areas when deploying systems.  We tried a variety of
carrying cases and finally discovered one that was very durable and
functional.  We liked this case because it provided a sling to protect
the laptop unit during tra nsit.  Other characteristics of a good
carrying case are:

A handle and strap for carryings.  External pockets and sleeves (nice
to carry plane tickets in).  A place for extra hard disks.  Storage for
cables.

Provide special laptop accessories in carrying case.  In addition to a
good quality carrying case, some additional accessories specific to
laptop machines come in handy.  Here is a list of accessories that we
found useful.  A telephone tester to identify digital jacks that can
destroy most PCMCIA modem cards.  A 20 foot phone cable.  A phone
splicer to connect cable from modem to 20 foot cable.  An extra modem
to phone jack connector (some cables break).  An extra battery pack
(separate charger is optional).  A digital-to-analog converter.


Register some laptops for Worldwide travel.  This lesson is only
important to businesses that engage in worldwide travel.  We discovered
a significant amount of paper work to file with U.S.  immigration for
each laptop that is used for international travel.  The number of
laptops needing to be regis tered depends on the number of people
traveling.  We found that one machine was able to satisfy most of our
needs.  We also have a set of plug adapters for use in foreign
countries.  Don't forget about the power supplies.  Some power supplies
handle dual voltage and some don't.  Check the hardware documentation
to be sure.

Consider a dedicated demo machine.  One of the advantages of a laptop
machine is it makes a convenient demonstration machine.  We have an IBM
ThinkPad 755CV dedicated to this.  It has 40MB of memory and the
overhead display screen comes in useful for giving demos when
projection hardware is not available.  Be sure to get a laptop that is
real snappy.  At the time of this writing, we found a Pentium 75
machine a good choice.


Workstation Lessons We deployed laptops as primary desktop workstations
about six months after we deployed the "travel pool".  We learned a
number of lessons from this deployment effort too.

Use a PCMCIA LAN adapter.  Choose a PCMCIA LAN adapter to make a
primary desktop machine portable.  The IBM ThinkPad 360 and 755 series
support both ISA and PCMCIA cards in the docking stations.  Although
its easier to implement and maintain an ISA configuration, the ISA
cards will not connect to a LAN when it is off the docking station.

Pre-install enterprise software locally.  Each enterprise or business
usually has a collection of application software approved for use.  In
some enterprises, much of this software can be run from the LAN.  This
presents a problem for the laptop workstation owner because he cannot
access the LAN- based applications when the laptop is not connected to
the LAN.  For this reason, we found we loaded most applications on the
laptop hard disk.  This can increase hard drive size requirements.

Pre-install travel remote connectivity software People who have laptops
as primary desktop workstations also need remote connectivity software,
just like the travel pool laptops.  We found that OS/2 TCP/IP 2.0
provided both LAN and dial-up connectivity.  This means the
applications that access enterpri se systems are exactly the same
whether connected via the LAN or dialed in through a SLIP (Serial Line
Internet Protocol) connection.  The only difference is how TCP/IP is
initialized.

Consider using a docking station, port replicator or similar device.
The IBM ThinkPad machines we deployed as primary workstations came with
docking stations.  These docking stations not only provided a place to
mount the laptop when at work, but also provided additional PCMCIA
slots, ISA card slots, additional bays for a hard drive and/or CD-ROM
drive (the ThinkPad 755C does not have a built-in CD-ROM), stereo
speakers and a security lock.  The docking stations also provide a way
to attach a full-size keyboard, conventional mouse, external monitor
and printer.  Since t hese devices connect to the docking station the
user does not need to re-connect the cables every time he mounts the
laptop.  The docking station makes the laptop more like a traditional
desktop workstation.

A port replicator is a poor-man's docking station.  It provides, a
subset of the function provided by the docking station.  The two PCMCIA
slots, the keyboard, mouse, monitor and printer ports are enough to
eliminate re-cabling.

The major advantage of docking stations and port replicators is they
provide additional functionality, save time by not having to recable
and eliminate problems that can arise with repeated connections.

Deploy laptops to people familiar with PC technology.  People
comfortable with desktop computers and who have used them for a
reasonable amount of time are candidates for a primary laptop
workstation.  People without much PC experience have more difficulty
using laptop workstations.  So you may need to provid e additional
education or support to the inexperienced PC users.

The deployment of laptops in the workplace appears to be a popular
trend.  Most likely, businesses will deploy the primary workstation
variety versus the "travel pool" variety.  In either case, no single
solution exists that fits all deployment situations. Of the lessons
presented here, we found the areas of security devices, power-on
passwords and anti-visus protection the most painful to learn.  We hope
that sharing our experiences will help others deploy laptops in the
workplace environment more effectively.


Kurt Ruby is a staff programmer with IBM in Rochester, MN.  You can
reach him at 507-253-1943 or via e-mail at KurtRuby@vnet.ibm.com.

Diane McBride is a Senior Procedures Analyst working with IBM in
Rochester, MN.  You can reach her at 507-253-3434 or via e-mail at
DMMcBride @vnet.ibm.com.

ELECTRONIC 'SCAPES 
Let's brew some more Java Mini-Course 2 
by Jose deJesus 

Welcome back.  This is the second part of a series of
mini-courses designed to teach you the Java for OS/2 development
environment and the Java language.  During the past few weeks, some of
you might have been grumbling that you wanted to see more, more, mo re!
And I don't blame you.  Java is a very exciting topic, and the more you
read about it, the more you want to know.

Okay, I admit it; January's article left you somewhat in mid-air.  But
then again, I never promised to teach you Java in a single article.
The purpose of the first lesson was to give you an idea about Java, and
show you how the compiler works--enough infor mation to get you
hacking.  I did mention some of the (VERY) basics, but when it was
starting to get good, POOF!  I disappeared.  Keep in mind that there is
only so much we want to put into an article (who says that can't change
though?  Please write to me with your suggestions at 73141,317
@compuserve.com).

In this lesson, I will cover operators, expressions, and statements,
and then move on to explain the object oriented programming paradigm.

Operators Java operators are symbols that let you form expressions.  An
expression is simply a combination of tokens--literals, identifiers,
operators--that yields a result.  For example:

x = (1 + 2 * 3);

The expression (1 + 2 * 3) uses the + (plus) and *(multiplication)
operators to return the value 7. The Java Integer operators include:

Type Operator Operation

Unary
       -       negation 
       ~       bitwise complement
       ++      increment 
       --      decrement 
       abs()   absolute value

Binary +       addition
       +=      addition 
       -       subtraction 
       -=      subtraction 
       *       multiplication 
       *=      multiplication  
       /       division 
       /=      division  
       %       modulo 
       %=      modulo 
       >>      right shift 
       >>=     right shift (propogate sign) 
       >>>     zero-fill right shift 
       <<      left shift 
       max()   maximum value of range 
       min()   minimum value of range

A unary operator works on a single operand, whereas a binary operator
applies to two operands.

Some rules apply to the type of result an expression produces:
Operations on integers never return a result of type character(char),
short, or byte.  If a variable x is declared byte, x + 2 would return
an integer(int).  If one of the operands is of type long, the result is
also long.  An expression produces a double-precision value, if at
least one of the operands is double-precision.  If the Java compiler
figures that the result of an expression will be outside an operator's
range, it reduces the answer to modulo the range of the result type.

There are also Boolean operators, which return either true or false.
These include:

Operator Operation 
!         negation 
&         logical AND 
&=        logical AND 
|         logical OR 
|=        logical OR 
^         logical XOR 
^=        logical XOR 
&&        logical AND 
||        logical OR 
>         greater than 
<         less than 
>=        greater than or equal to 
<=        less than or equal to 
==        equality 
!=        inequality 
?:        ternary if, then

Notice that operators follow a precedence order.  Java evaluates
(1 + 2 * 3) as being 7, not 9, because the multiplication operator has a
higher precedence than the addition operator.  Parentheses can override
these rules.  For example, to get 9, you could w rite the above
expression as (1 + 2) * 3. In this case, parentheses will force the
addition to occur before the multiplication.  The precedence order of
Java operators from highest to lowest is as follows:

. 
[] 
()
++ -! ~instance of 
* / % 
+ - 
<< >> >>> 
<> <= >= 
== != 
& 
^ 
| 
&& 
|| 
?:
= += -= *= /= &= |= ^= %= >>= <<= >>>= 
,

An expression with two operators of equal precedence (as in the * and /
operators above) evaluates from left to right.  Thus, the expression
(10 / 20 * 4) returns 2, not 0.125.  Consequently, we say that the *
and / operators have left to right (as opposed to right to left)
associativity.  The associativity for most Java operators is left to
right.

Statements Java supports simple and structured statements.  Simple
statements include declarations, assignments and function calls, such
as:

int x;                      // declares x as an integer variable
x = (1 + 2 * 3);            // assigns the value 7 to x 
MySub(x);                   // passes x to the function MySub

Structured statements can be compound or of control flow.  Like in
C/C++, a Java compound statement is simply code enclosed within curly
braces:

{ 
<code>
 }

=====================================================================
Note:  From here on, the convention <code> will mean program code.
I'll be using it in some examples to avoid showing you actual code that
might be irrelevant to the discussion at hand.  You'll also see
<statement>, <boolean>, <expression>, and <label>, to indicate
statements, boolean results, expressions, and labels, respectively.
=====================================================================
Control Flow Statements Control flow statements are used to control the
way a program flows.  Except for the goto statement, Java's control
flow statements can be conditional, repetitive, transfer, and
exceptions.

=====================================================================
Note:  Although goto is a reserved word, Java does not support the goto
statement.  Instead, Java programmers use labels with reserved words
such as break and continue.
=====================================================================

Java's conditional statements include if, if-else, and switch
constructs.  Basically, their format is as follows:

/** Format of the if statement **/ 
if <boolean> 
<statement>;

<code>

If <boolean> evaluates to true, <statement> executes.  Otherwise, the
program skips <statement> and continues executing the rest of the code.
<statement>, of course, can be either simple or compound.  Here is an
example of an if statement:

if (x == 7) 
   { 
     System.out.println("Your answer is correct.\n");
     System.out.println("Welcome!");
    }

The if-else statement has the same format as the if statement, but with
a twist:

/** Format of the if-else statement **/ 
if <boolean> 
   <statement1>; 
else
   <statement2>;

<code>

If <boolean> evaluates to true, <statement1> executes, and the program
continues with <code>.  Otherwise, <statement2> executes before the
program continues with <code>.  For example:

if (x == 7)
   { 
    System.out.println("Your answer is correct.\n");
    System.out.println("Welcome!"); 
    } 
else  
   { 
     System.out.println("Your answer is incorrect.\n"); 
     System.out.println("Arrest this intruder!");
   }

A switch statement evaluates an expression, and tries to find a
matching case.  When one is found, it executes a corresponding
<statement>.

/** format of the switch statement **/ 
switch (<expression>) { 
  case <expression>:  <statement>; 
  case <expression>:  <statement>; 
  ...
  default:  <statement>; 
}

Notice that the default keyword is used to handle all other cases.
Here is a code example of the switch statement:

switch (x) 
  case 1:  System.out.println("one"); 
  case 2:  System.out.println("two"); 
  case 3:  System.out.println("three");
  default: 
           System.out.println ("x is not equal to one, two, or three");
end case;

For repetitive or iterative statements, Java supports for, while, and
do-while loops.  Their formats are as follows:

/** Format of the for loop **/ 
for (<expression1>; <expression2>;<expression3>) 
    <statement>;


The for loop causes <statement> to execute a fixed number of times
until <expression1> reaches the value of <expression2> through
incremental levels of <expression3>.  Here is an example:

for (i=0; i<10; i++) 
    { 
     System.out.println(i); 
     System.out.println("Mississippi\n"); 
     }

The format of the while loop is:

/** format of the while loop **/
while <boolean> 
   <statement>;

The while loop continues to execute <statement> as long as <boolean>
remains true.  For example, to increase and display the value of x as
long as it is less than 7, the loop would look like this:

while (x < 7) 
     { 
      x = x + 1; 
      System.out.println(x); 
      }

The do-while loop works similar to the while loop, except that, unlike
while, do-while executes the body of the loop (<statement>) at least
once:

/** Format of the do-while loop **/ 
 do <statement> 
 while <boolean>;

This type of loop is known as a post-test loop because it tests the
<boolean> expression after executing <statement>.  A loop, such as
while, that tests the condition before executing <statement>, is known
as a pre-test loop.  The following do-while example increases x by 1 at
least once, whether or not the value of x is less than 7:

 do 
  { 
    x = x + 1; 
    System.out.println(x); 
   }
   while (x < 7);

The transfer statements include break, continue, and return.  Their
formats are as follows:

/** format of the break statement **/ 
  break <label>;

/** format of the continue statement **/ 
  continue <label>;


/** format of the return statement **/ 
  return;

Any statement in your source code can have a label, which is simply an
identifier followed by a colon, placed before the labeled line.  Break
and continue statements use labels to transfer control within or out of
loops.  Functions use the keyword return to transfer program execution
to the line following the function call.

Exceptions Exception handling is another mechanism by which you can
affect a program's flow of execution.  An exception is an abnormal
error or condition that occurs while a program is running.  Although
Java provides standard exception-handling services, you can impl ement
your own exception handlers to catch and handle runtime errors.  Since
exception handlers are derived from the Exception system class, they'll
be covered in a later lesson, once you're up to speed with classes.

The OOP Paradigm If you want to understand Object Oriented Programming,
think in terms of objects, just like you do in the real world.
Consider, for example, a car.  A car is an object.  To drive it, you
need to use the transmission, accelerator, steering wheel, brakes, etc
. These devices interface between you and
the car, and, therefore, you must know how to use them.  However, you
don't have to know how they work internally to be able to drive the
car.

Software objects work the same way.  If the car were a software object,
to use it, all you would have to know is how to interface with it.  The
way you do that is through the object's public methods.  Methods are
simply routines that are tied to the object.

Say the car's fuel is a variable called fuel.  In a traditional
programming language, accelerating the car would cause a call to some
function, say accelerate, that affects the value of fuel.  Fuel could
be a global variable that other functions (i.e.  startcar) could easily
access.  While this sounds like a clear-cut plan, an OOP enthusiast
would tell you that there is a problem with this picture:  your
program's data and the functions that act upon it are not together.

In the real world, it would mean driving a car that uses gas from an
unattached fuel tank, if you could imagine that at all.  Of course, in
this scenario, a gas meter would be useless since other events (not
just the car's) could affect the contents of the fuel tank.  OOP
dictates that you should model your programs more after the real world.
A car object, for instance, should encapsulate its fuel variable, and
the functions that will affect it (i.e.  startcar, accelerate) into a
single data type.  This chara cteristic, known as encapsulation,
provides several benefits:  Data Protection--since only the object's
methods can affect its data, code is cleaner and easier to debug.
Information Hiding--an object has a public interface to the outside
world, but it can also have private methods that the OO programmer can
change without affecting any programs that depend on those objects.
For example, the brake mechanism on my car can c hange without
affecting the action I take to stop the car (i.e.  stepping on the
brake pedal).

An object is actually an instance of a data type known as a class.  A
class is a template for an object, the same way a data type is a
template for a variable.

Now consider another important characteristic of OOP:inheritance.  If I
tell you that I own a porsche (yeah right), you would automatically
think of it as a sports car, and expect it to have the parts that every
car has (i.e.  an engine, an accelerator, bra kes, etc.).  The link
between car and type of car is implicitly there.  So it would be
foolish for someone to ask if a porsche has brakes.  Of course, it's a
car!  OOP allows you to do the same with software.  You can define
classes in terms of other classes, and that way build a hierarchy.  For
example, a porsche would fit nicely under the following classes:

-----------Car----------------------- 
|               |                   | 
Sports Car      Luxury Car         ...
|               | 
|               | 
Porsche        Lexus 
(etc)          (etc)

In this hierarchy, Porsche has all of the characteristics of a sports
car, which in turn has all of the characteristics of a car.  In Java,
Porsche is a subclass of Sports Car, which is a subclass of Car.
Conversely, Car is as a superclass of Sports Car, w hich is a
superclass of Porsche.  Inheritance is also transitive; if Porsche is a
subclass of Sports Car, and Sports Car is a subclass of Car, then
Porsche is a subclass of Car.  In this case, Sports Car is the
immediate superclass of Porsche.  The actual code would look something
like this:

class Car {

/* class body */
     }

class SportsCar extends Car {

/* class body */ 
   }

class Porsche extends SportsCar {

/* class body */ 
   }

With inheritance, I can define a whole new series of cars without
having to redefine the methods and variables that apply to all cars.
These will automatically be available to my new class since I can
extend the features already found in Car, or in one o f its subclasses.
The point to get here is that, through inheritance, I am reusing and
extending code that already exists.

Java only allows single inheritance.  Other OOP languages, such as C++,
support multiple inheritance, where a subclass can have more than one
parent class.

In Java, the mother of all classes is called Object.  Every class,
except Object, must have an immediate superclass.  When you don't
specify a superclass, Java implicitly assumes the superclass Object.
Thus, the class declaration for Car is the same as if it were written
as:

class Car extends Object {

/* class body */ 
  }

Both forms are acceptable to Java.

Different classes in a hierarchy can share the same method names,
through a technique called polymorphism.  You'll see polymorphism at
work as you start working with Java classes.  Classes are the lifeblood
of Java programs.  Becoming familiar with OOP concepts is the first
step toward learning Java programming.


A Peek at TCP/IP By Marc-Arthur Pierre-Louis


TCP/IP is an evolving standard to link heterogeneous networks into a
global network not bound by hardware nor operating systems.  It is a
protocol suite bearing the name of its two most famous protocols:  IP
or Internet Protocol and TCP or Transmission Control Protocol.  The
TCP/IP standard works in between networks to allow a reliable and
unified communication among them.  Thus, the term inter-networking
lends itself quite well to the scheme.

In the wake of the computer revolution and the advent of the Internet,
the importance of a global network has significantly increased.  As
this happens, TCP/IP grows significantly in stature.  Just as one does
not need to know much about the make up of a ca r to enjoy riding in
it, it is not necessary to understand the underpinnings of the protocol
suite in order to partake in the benefits that it offers.  After all,
the suite provides a great deal of abstraction, shielding users from
the particularities of hardware and networks and applications
programmers from having to know much about specific implementations of
the protocols.  However, an informed user is bound to benefit when
he/she knows the history, evolution, implementation, and direction of a
product.

History & Evolution

Although achieving popularity just in the past few years, the standard
is not very young when compared to the life cycle of most products in
the computer industry.  Its origin dates further back than the age of
the Personal Computer.  It was the result of brilliant research and
clever forethought.

Around the mid 1970s, the Advanced Research Projects Agency (ARPA) of
the U.S.  Department of Defense began conducting research on connecting
networks.  By 1977-1979, the communication protocols for a global
network started taking more definitive shape.  As a result of that
work, ARPANET, the backbone of the global Internet, was formed in 1980.
The Internet was a result of ARPA converting from previous protocols to
TCP/IP.

Another move that assured the success of the protocols was influencing
universities to adopt the protocols.  ARPA achieved that by making
procurement of an implementation relatively inexpensive.  The
University of California was funded to include the protocols in the
Berkeley Software Distribution program.  This program reached virtually
the whole university community.

By 1986, the National Science Foundation entered the arena and began
funding a Wide Area Network (WAN) known as the NSFNET, which became the
second backbone of the global Internet.  The global Internet, using
TCP/IP protocols, thus began a phenomenal growt h that has reached
exponential dimensions.  Today, the global Internet spans dozens of
countries with thousands of networks and millions of users.

Implementation The implementation of TCP/IP follows the layered model
proposed by OSI on protocols.  But as opposed to the seven layers in
OSI model, the TCP/IP uses a four-layer model.  Designed to run on top
of networks, it follows the same principle of networks.  Data is made
into packets that are sent over the communication channels.  The
following is an overview of the layers:

Application 
Transport 
Internet 
Network


The Network layer is responsible for communicating with the specific
network on which the TCP/IP software is running.

The Internet layer deals with communication from one machine to the
other.  It routes requests to specific routers determining the initial
path the request should follow and encapsulates requests in the format
specified by the protocol.  It also sends, rece ives, and handles error
control messages as appropriate.  This is the fundamental layer in the
protocol suite.

The Transport layer's role is to regulate the flow of communication
between applications.  This is known as end-to-end communication.
Application programs send streams of data to the transport layer, which
sends them to the Internet layer.  When the Intern et layer sends data
back to the Transport layer, the latter determines the application
destination and sends the data.  It is possible, depending on the needs
of the application, to perform rigorous error checking at the transport
layer.  Several protocols may also be defined at this layer.

The Application layer is the highest.  Applications choose what kind of
transport service they want, based on their needs, and send the data in
the required format to the transport layer.

Protocols

TCP/IP defines several protocols in addition to IP such as:  Internet
Control Message Protocol (ICMP) Address Resolution Protocol (ARP)
Reverse Address Resolution Protocol (RARP) User Datagram Protocol (UDP)
Transmission Control Protocol (TCP).

The IP Protocol The Connectionless Packet delivery service is the basis
of all the other services offered by the protocols.  It delivers
packets from one machine to another based on the address specified in
the message it is carrying.  This service is unreliable because of the
separate routing that each packet undergoes.  This results in packets
being lost, not delivered on time, or not delivered in the correct
order.  Although unreliable, the Connectionless Packet Delivery Service
is very efficient because of its close inte raction with the underlying
hardware.

To overcome the reliability problems associated with Connectionless
Packet Delivery service, IP incorporates the Reliable Packet Stream
Transport Service.  This service establishes connections between two
applications on separate computers and streams of data are sent
between them.  The streams of data will still be broken into small
pieces and sent one at a time, but Reliable Packet Stream Transport
services add robust error checking and recovery to ensure a reliable
system.

The IP protocol is known as a connectionless service.

Three basics things are necessary besides the underlying network:

An address where to send information (an IP address) A piece of
information to send (Datagram) A routing agent to forward the
information (Router)

IP Addresses:	

The address to send the information is known as an IP address.
Different from the network address, the IP address is a 32-bit entity
that encapsulates information about the network and the host (the
computer running the software).  IP addresses are cleverly divided
into five classes, where classes are assigned according to the needs
and size of a specific network..  The following could represent valid
IP addresses:  					
     8.16.9.45; 239.255.178.190; 12.12.89.90

The address is set up so that the network address can be extracted
rapidly.  For readability, the address is divided into four decimal
numbers each representing one byte of the four-byte number.  A dot
separates each decimal number.  Hence, the following 32- bit IP
address:
     11110000000110000101010000000

is written as:  					
     254.3.10.128

These IP addresses must be unique.  No two hosts can share the same
address.  The analogy is that your mail would be mixed up if you had
the same address as your next door neighbor.  It is possible for a host
to have several IP addresses, in the case of a mu lti-homed host, and,
as we will see later , in the case of routers.

Datagrams

The Datagram is the piece of information that the IP protocol sends.
It is the unit of transfer.  No matter how large or small the data, it
arrives at the destination as one or several datagrams.  It is analgous
to network packets.  The format of the datagram follows:

Datagram Header	 Datagram Data Area


The datagram header is further divided into several 32-bit fields that
may contain one or more sub-fields.  The following terms are contained
in the header.:


The version field is used for compatibility reasons.

The length of the header lets the receiving side know where the data
starts.  The header is not of fixed length because fields may increase
or decrease in size.

The Service Type field indicates how the datagram should be treated.
This field contains several subfields that indicate the importance of a
particular datagram.  The routing software looks at these fields and
may make routing decisions based on their valu es.

The Total datagram length is used for administrative reasons.  Since
the datagram is made up of a header and subsequent data, the data
length can be quickly computed by subtracting the header length from
the total length.

The Identification field is chiefly used in the case where datagrams
exceed the size of network frames.  In such cases, the datagram is
broken in several fragments that are reassembled at the receiving side.
The ideal situation would be that each datagram be sent in one network
frame but because each network has its own frame size it is sometimes
necessary that datagrams be broken down into parts.  When fragmentation
is necessary, all the fields of the second 32-bit word of the header
(Identification, Flags , and Segment offset) are used for that purpose
and for re-assembly at the receiving side.

The Time to Live field specifies how long the datagram is to remain in
the system.  Past that time, the datagram is discarded.  This is to
prevent undeliverable datagrams from remaining forever in the system.
Consequently, unusual delays will cause datagra ms to be lost.

The Protocol field specifies the high level protocol used to format the
data.

The Header checksum is used for error control on the datagram header.
Note that there is no checksum for the data.  Higher level protocols
ensure the correctness of the data received and sent.

The IP Source and Destination addresses need not be explained.

The IP Options field is primarily used for network testing and
debugging and the filler, is used for alignment.

Based on the needs of higher level protocols, the data portion of the
datagram may be broken down and may contain its own header and data
correction fields.  Basically, any information flowing between systems
that relies on the IP protocol will have to comply to the above data
layout.

The Routing Agent

Routing is at the heart of the IP protocol.  It determines which path
the datagram will take.  Given a network, there must be an entity that
makes the decisions to forward the information based on the IP address.
For instance, if the IP address lies within the local network then no
outside path will be chosen.  We thus distinguish between direct
delivery and indirect delivery.  In direct delivery, the datagram is
sent to a machine that is in the same physical network.  In such cases,
no router is involved beca use the capability to converse among the
different machines in the network is inherent to that network.  All
that is needed is to fit the datagram in (a) network frame(s), bind the
IP address to the hardware physical address and send the frame(s)
directly.

Indirect delivery is for sending datagrams across different networks.
This necessitates the services of a router.  A router is a computer
that connects physical networks.  Without them, inter-networking would
be impossible.  Hence, a host (machine running on the network) will
send the datagrams to the router that will choose a path based on the
IP address.  Obviously, a router must be known to the physical networks
it connects and will have IP address for each network.  Routing may
depend on using routing tabl es that store destination routes.  Since a
system must store a tremendous number of routes, using these tables
consumes a great deal of space.  A method called next hop routing,
reduces the required space by maintaining just enough information to
reach the next router, which in turn will worry about finding the next
hop.  It is advisable that routers be dedicated machines; but, host
systems can perform routing.  The latter case can lead to some
hardships for the host machines.

Routers must know the physical addresses of all the machines en route
to the destination because the datagram only has the IP address and not
the physical network address.  This gives way to a couple of protocols
known as ARP and RARP.  In ARP or Address Resolution Protocol, the
router will broadcast a frame containing the IP address and the
physical address of the sender to determine the physical address
associated with the IP address. The owner of the IP address will reply
with its physical address and the datagram can thus be sent to it.
The RARP (Reverse Address Resolution Protocol) is the converse of RARP
and is used in the case where the physical address is known but not the
IP address.  Normally ARP and RARP are not considered as part of the IP
prot ocol although they are used by it.

SLIP & PPP Two other protocols worth mentioning are extensions of the
IP protocol to provide inter-networking functionality to single
machines not formally part of a physical network. SLIP stands for
Serial Line IP and is used as a protocol to frame IP datagrams over
serial lines.  PPP stands for Point to Point Protocol and is also used
for serial lines.  These protocols add another layer to the four
conceptual layers of TCP/IP.  This layer, called the Intranet layer,
lies below the Network layer and is responsible for routing between the
Network interface layer and the hardware devices.

Conclusion To recapitulate, TCP/IP consists of four conceptual layers:
Network interface, Internet, Transport, and Application.  There may be
an additional layer called the Intranet layer, with SLIP or PPP.  The
Connectionless Packet Delivery Service that makes up the IP protocol is
the basis of all other protocols and services.  The IP protocol uses IP
addresses and datagrams to send information that may travel across
several networks to reach its final destination with the help of
routers.  Depending on the higher pr otocols used, the final
destination may apply a tremendous amount of processing to the data to
ensure accuracy and reliability.  In short, TCP/IP is the networking of
networks or Internetworking.  It began in the mid 1970s and is still
evolving.

The term technological leap applies when one attempts to summarize the
possibilities that the protocol affords to applications and other
protocols lying above it.  This is quite a leap from its humble
beginning.  TCP/IP is achieving fame and renown in the homes where SLIP
and PPP are extending the reach of the Internet to anybody having
access to a modem, computer, telephone line, and most importantly some
software package implementing TCP/IP.


FEATURES 
Lotus Notes Release 4.0:  Another technological leap 
by Jeri Dube

From the beginning, Lotus Notes has epitomized a technological leap.
With Notes, Lotus practically invented the idea of groupware.
Groupware breaks down barriers to using automation for effective
communication, collaboration and coordination.  Regardless of computing
platform or location within an organization, people can easily access,
track, organize and share information.  Notes accomplishes these feats
with distributed document storage, integral messaging, replication,
security and other system level services.  Notes also offers a
complete application development environment.

Lotus continues to innovate and improve on the original Notes product.
Some of the principles in developing the fourth release were to
increase manageability, availability and scalability to support
enterprise-wide and inter-enterprise mission critical applications.
The functional improvements guided by these principles can lead to
significant decreases in cost of ownership for Notes customers.

With release 4.0, Notes customers can decrease the number of servers
required to support their end users.  Where Notes Release 3.x supported
from 100 to 200 users, the fourth release can support up to 1000
active, concurrent users.  Architectural improvemen ts in release 4.0
give Notes the ability to support symmetrical multiprocessing systems
with up to six processors.  Consequently, each server can handle more
simultaneous sessions and has increased replication capacity.

Lotus implemented several clever ideas to improve replication
throughput and performance and to offer unparalleled information
distribution capability.  Each server now has multiple replicators that
perform bi-directional synchronization no matter where a replicate
database is located.  Rather than replicating an entire database, Notes
release 4.0 only replicates the fields that have changed.  They also
implemented a Notes server PassThrough, whereby mobile users can access
multiple servers with a single t elephone connection.

The PassThrough feature has several advantages.  Users can easily get
to any information regardless of where it is stored.  PassThrough also
eliminates complexities associated with managing different network
protocols.  System administrators do not need to configure systems
individually to conform to LAN or WAN infrastructures.  They can manage
any server remotely despite disparate protocols.

Notes 4.0 eases administrative tasks with a number of new intuitive
tools.  These new tools centralize and streamline administration across
an enterprise.  These also reduce ownership costs by reducing the
number of administrators required to manage a Notes enterprise and
decreasing the amount of time and money it takes to train them.  The
new tools include:

Administrative control panel
Server console panel 
N&A and ACL management tool  
Delegated administration 
Server configuration record

The new administrative control panel and the server console panel,
both, offer an easy-to-use interface for administrators.  The control
panel makes it more intuitive to manage users and groups as well as to
view message logs.  Administrators can use the control panel to manage
several servers simultaneously from a single console.  The server
console panel groups functions and consolidates information into a
single panel with a simple graphical button for a number of
administrative activities.

The N&A and ACL management tool automates the user administration
process by making the necessary changes to all server access lists,
groups, and access control lists (ACL) throughout the Notes environment
in a secure, reliable manner.  This tool reduces the amount of time
administrators spend maintaining and updating user accounts and ACLs.
It also provides greater security for the entire Notes environment.
When an administrator deletes a user from one list, the user is deleted
from the corresponding lists on all servers throughout the enterprise.

Both the delegated administration feature and the server configuration
record aid in centralizing administrative tasks.  Delegated
administration lets administrators assign certain administration
privileges or roles yet maintain centralized control.  This leads to
more secure server operations, minimizes error and increases
administration efficiency and productivity.

The server configuration record allows administrators to make changes
to servers throughout the network from any network server.
Administrators can make global changes to enterprise server
configurations or change only a subset of the servers.  This too l
makes fine tuning or troubleshooting much easier tasks to accomplish.

Notes release 4.0 provides complete Internet integration.  Users will
be able to do server-to-server as well as server-to-client integration
over the Internet.  This process will be much easier to set up and
manage, with IP addresses appearing as any other Notes server in a
replication dialog box.  Notes 4.0 servers will support anonymous
clients so organizations can provide a Notes server accessible to any
Notes client using the Internet.

A key part of release 4.0 Internet integration is the InterNotes Web
Navigator, which lets Notes manage Web documents without altering their
look or content.  The agent builder in release 4 will allow users to
automate tasks such as scanning targeted Web sites for new or modified
pages.  The agents then can send daily summaries to a user's e-mail.
Release 4.0 offers advanced Web browsing functions such as the ability
to forward a Web page while browsing.  Another browsing function lets a
user forward the co mplete route he used to get a particular Web site
to another user.  Notes release 4.0 can also limit access to the Web.
This ensures that employees use the Web only for productive purposes.

The latest release provides many enhancements for Notes end users.
Lotus has improved the user experience for key user activities
including finding information, viewing information and creating and
modifying information.  Other improvements from the clien t perspective
are enhanced mobile support as well as added functionality to the mail
system.  Lotus has also enhanced the development environment making
Notes applications easier to write.

The Notes release 4.0 server currently supports the OS/2 platform.
Lotus intends to ship client support for OS/2 within 90 days of the
server's January shipment date.


OS/2 Warp Server - A giant leap for LANkind 
by Richard Hawes


OS/2 Warp Server will be more than just a bundling of existing
technologies.  Its enhancements and the integration itself will
represent technological leaps in server function.  For example, the
OS/2 Warp Server Tuning Assistant will automatically set resource
parameters based on the complete range of components in use, including
System Management and Remote Access services.

OS/2 Warp Server is the first of a major set of server products IBM
plans to release as part of a project called Eagle.  Servers in the
pipeline include an Internet Server, Database Server, Transaction
Server and others.  In addition, IBM intends to make th e same high
quality products and implement them on other platforms besides OS/2.

OS/2 Warp Server will combine many sought after server features in one
integrated package.  Despite covering all the many, varied components,
the installation program will attain new levels in ease of use.  Most
parameters can generate defaults, if required. The interface features 
will be more appealing, informative and intuitive with the use of graphics. 
For instance, the advanced install option includes a configuration 
panel with an easy-to-understand data format.

Both the U.S.  and National Language versions of OS/2 Warp Server are
due for launch at the end of February, 1996.  And it is likely to be
aggressively priced.  Comparable state-of-the-art functionality on any
other server platforms is either unavailable or overpriced.  The
impressive list of components follows.


OS/2 Warp OS/2 Warp brings all the advantages of OS/2 Crash Protection,
preemptive multitasking and so on.  The Internet Access Kit from OS/2
Warp Connect and the other BonusPak applications will also come with
OS/2 Warp Server.


Adapter and Protocol Services - Connecting to the world.  OS/2 Warp
Server will contain a variety of protocol stacks including TCP/IP,
NetBIOS and IPX.  We've extended built-in hardware support to include
more new adapters making installation much easier especially with the
new adapter sniff facility within the i nstallation program.  The
TCPBEUI protocol, which allows LAN Server and other NetBIOS
applications to use TCP/IP for wide area networking, has been enhanced.
It now supports over 1000 sessions across a single physical network
adapter.


File and Print Services -Network administration as easy as
drag-and-drop.  The file and print services are based around the
award-winning LAN Server 4.0.  The services will provide the normal
server functions such as network application access and printer and
modem sharing.  In addition, the server can synchronize passwords
through Network SignON Coordinator/2.  This will facilitate easy
password management for multiple host and LAN systems.

A new NetWare Gateway functions will further extend OS/2 Warp Server's
file and print services.  This function will establish a session with a
designated NetWare server and then in turn will share the resource with
OS/2 Warp Server users.  With this gatewa y connection, the users can
access NetWare resources using standard OS/2 Warp Server protocols.
This means the installed client software will not need to include
NetWare support.  This facility has been available in IBM's LAN Server
since the NetWare client has been available.  In OS/2 Warp Server, the
installation and configuration process will integrate it.  The same
gateway ability will allow clients to use any file or printer resource
that the OS/2 Warp Server can access such as NFS, AS/400, or Client 
Access/400.

Other enhancements to the server facility will include an improved
tuning utility, better messaging, an improved graphical interface for
administrative tasks and a new facility to aid installation of clients
from across the network.  The server will support OS/2, DOS, Windows
3.x and Windows 95.

Remote Connection Services -- LAN connectivity wherever you are.

Remote workstations will have access to OS/2 Warp Server.
Consequently, the network through the server can connect to various
kinds of switched and non-switched links such as telephone connections
and ISDN.  It will be based on and fully compatible with the IBM LAN
Distance products, including the remote client provided in OS/2 Warp
Connect.  Support will include NetBIOS, TCP/IP, Lotus Notes and
cc:Mail, NetWare and DB2/2 Requesters -- in fact any application that
complies with the network driver interface specification (NDIS) or uses
an NDIS protocol will be supported.  The Remote Access Service
effectively will behave as a bridge into any LAN segment with which it
is connected.  It will connect to two remote LANs therefore acting as a
bridge between them.  This service will also act as a 'virtual LAN' by
linking two remote workstations.


System Management Services -- It's great to be in control.

Based around the very successful Netfinity system management utilities,
OS/2 Warp Server will provide a wide range of system management
services.  Some of the client monitoring and management functions will
be software and hardware discovery, license manag ement, and software
distribution based around Netview DM/2.  The remote workstation control
and resource monitoring features will be a real bonus to technical
support and help desk teams.


Backup and Restore Services -- Disaster?  What disaster?  

Personally Safe and Sound has won awards for the drag-and-drop 
approach to maintaining backups.  With incorporation in the 
OS/2 Warp Server, you will be able to backup data sets to practically 
every type of device, even across a network.  Support for the 
ADSTAR Distributed Storage Manager product (ADSM) means network 
administrators will be able to integrate OS/2 Warp Server backups 
into an Enterprise-wide backup procedure across numerous platforms.  
ADSM is available on OS/2, AS/400, MVS and numerous UNIX 
incarnations.  The restore-enabled boot diskettes will make disaster 
recovery comparatively simple.


Advanced Printing Services -- Publication perfection

Beyond the usual printer sharing of the File and Print Services, OS/2
Warp Server's Advanced Printing Services will employ the latest
advances in printer control technology to provide a new level of
functionality.  It will be possible to make use of bi-directional
printer interfaces, giving administrators and users concise information
about the status of printers.

Another great advance is the incorporation of data stream conversions.
These will allow the printing of PostScript documents on HP-PCL or
other printer types such as IBM host printing through AFP (Advanced
Function Printing).  Users will be able to print PostScript documents
on host printers.  That means users could be able to print a Worldwide
Web document at 300 pages per minute!  Speaking of saving time -- OS/2
Warp Server will save even more time on the client side because when
you add or change server printers, you will no longer have to install
new printer drivers on the client.


TCP/IP Services - Intelligent, Integrated Internetworking

The TCP/IP provision in OS/2 Warp Server and clients will prove IBM has
leapt into the future with TCP/IP support on the PC platform.  The
TCP/IP services will really minimize administrative tasks.  Imagine
being able to move a TCP/IP-based client to a new office in your
domain, without requiring a reconfiguration even if the move includes a
new subnet.  The client can keep his hostname and will not have to
worry about name server addresses, subnet masks or even the address of
the local mail server.

Connection to the wider world will become another simple task with
Dynamic Host Configuration Protocol (DHCP) and Dynamic Domain Name
Services (DDNS).  DHCP introduces a new level of availability into a
TCP/IP network.  With multiple DHCP servers existing i n the same
network -- if one fails, the client can automatically obtain DHCP
services from another.  Besides IBM's client software, DHCP is
compatible with Windows 95 and Windows NT clients.  Plus, a
drag-and-drop configurator makes server configuration much easier.
For more information on DHCP and DDNS see Introducing Dynamic IP for
OS/2 Warp Server in our December issue located at
http://www.austin.ibm.com/pspinfo/m7dynip.htm.

The Presentation Manager Single Input Queue Fix 
by Aidon Jennery

The Single Input Queue Fix (SIQFix) enhances performance in OS/2's
Presentation Manager focus switching system.  The focus switching
system brings windows to the foreground and opens or closes windows.
In short, this system ensures that input goes to the window that the
user intended.

Before the SIQFix, the focus switching procedure required, in all
cases, that the window with focus respond to messages before the new
target window receives its messages and gains focus.  If the thread for
the window losing focus is busy doing other work , then the PM user
interface would hang until that thread could process those messages.
This could hang the user interface for some time while the user waits
or hits Ctrl-Esc.  Ctrl-Esc would set a timer that would invoke "The
application is not responding" dialog and give the user the
opportunity to terminate the offending application.  Sometimes, for a
variety of reasons, the application would not terminate and the user
would be left with an inoperable system due to this suspension of the
user interface.

The SIQFix attempts to alleviate this problem by allowing a set amount
of time for the window losing focus to respond to those messages.  That
is, the thread would be dedicated to receiving the lose focus messages
before going off to another task.  If the thread processes the messages
within that time, then focus switching occurs as before.  However, if
the window does not respond, and a user operation causes the focus
switch (Ctrl-Esc, Alt-Esc or a mouse click on another window), then the
message queue fo r that window is marked as bad.  The system literally
rips away focus from the bad window and gives it to the new focus
window.  If the thread starts processing messages from its bad queue,
then the queue is marked as good and the previously bad window
processes the messages that it would have in a normal loss of focus.

The SIQFix also provides a few other advantages.  It will prevent the
user from trying to switch focus to a window whose queue has been
marked as bad and will also repaint the "bad" windows with an inactive
border and, in some cases, with the default wind ow color.  In
addition, if the user tries to close an application from the task list
that has been marked bad, then a DosKillProcess forcefully terminates
it.  Before the fix, PM would just send it a WM_QUIT message (the
message used to close application s that are not marked as bad).

The SIQFix will be available on future OS/2 product fixpaks.  By
default, the SIQFix is set to off so the system behaves as before. You
can activate the SIQFix by adding the following statement to
CONFIG.SYS.

SET PM_ASYNC_FOCUS_CHANGE=ON

This will enable the SIQFix giving a window 2 seconds (2000 ms) to
respond to the lose focus messages.  To set a custom time-out (for a
time-out of 5000 ms or 5 seconds as an example), use the following
statement:

SET PM_ASYNC_FOCUS_CHANGE=ON 5000

In this manner, you can tune the SIQFix to the system load and normal
application responsiveness.  To disable the SIQFix, just remove the
line from CONFIG.SYS or change it to:

SET PM_ASYNC_FOCUS_CHANGE=OFF

As development of OS/2 continues, we will enhance the SIQFix to address
other types of user interface hangs to provide a more responsive,
robust and reliable operating system.


Aidon Jennery is an Advisory Programmer on IBM's OS/2 Window Management
Development team.  

Preview:Leaps in Lotus technology make team work easier 
by Jeri Dube

The team computing features coming soon in SmartSuite 96 Edition for
OS/2 Warp embody technological leaps that can make teams work more
efficiently and more effectively.  Lotus development designed the team
computing functions with an understanding of how teams need to work to
make businesses run better.

Life Before Team Computing Technology Just about everyone who has
collaborated on a project with another person can tell a tale of how
the lack of technology inhibited the smooth completion of the work.
The ensuing example shows the difficulties encountered by a trio of
writers developing a white paper.  Two of the people worked at the
same location; the third worked 2000 miles away in a different time
zone.

To start the paper, each of the three team members developed several
sections.  Once they completed the first draft of the sections, the
first collaborative task was to put all them together in the right
order.  Since the team did not agree on the order, they saved two
versions in separate files.  They figured that their initial review
process would clarify which flow worked better.  This decision forced
them to review and edit both versions.

To avoid confusion or at least diminish it, they decided to use hard
copy for the review and edit process.  This way they would know who
made what comments to each version, by the handwriting.  Using hard
copy required them to fax between the two locations. Once the 
team agreed to changes for each version, they added them to 
the soft copy documents.  Finally after further discussions 
and arguments, they cut and paste pieces of both versions together 
into their first complete draft of the paper. They put this draft 
into a third file just in case they changed their minds and
wanted to use something from either of their first two original
versions.

With that work (painfully) accomplished, they initiated a management
review process.  Some of the managers made changes to the soft copy and
others just wrote their comments on hard copy.  The final team task was
to incorporate everyone's additions, delet ions, suggestions, and
changes.  To do this, one member of the trio gathered the submitted
hard copy and printed hard copy of all the soft copy input.  Then, one
by one, he went through all the hard copies incorporating the changes.

Although tedious, this went fine until he realized that two reviewers
made different suggestions to change the same paragraph.  Faced with
this dilemma, he had to get the rest of the threesome together to
decide which suggestion to use.  Since this happene d at the end of the
day, his team members, located in the earlier time zone, had left.  The
next morning they finally finished the paper.  With the final draft
saved, printed and distributed, the group was able to throw out the
huge pile of paper that had accumulated and delete the multiple down
level files.


The Universal Hurdles and Benefits of Team Work This less-than-elegant
scenario is not unique to teams developing white papers.  The same
hurdles can occur doing any collaborative project.  They can crop up
when a team has to consolidate budget input from multiple departments
into one meaningful spreads heet or create a presentation with a
consistent look and feel from several presentations all with different
styles.  They occur whenever people need to integrate their unique
offerings into united, consistent output.  These problems are
universal.  They do not discriminate based on industry, company size or
team member personality types.

Fortunately, the benefits of teams working together are also universal.
Collaboration may be the soul of today's business world.  Understanding
this, Lotus analyzed how teams work so they could provide solutions
that allow companies to reap the benefits o f team work more easily.
From their analysis, they found a number of processes cumbersome and
complex, including:  Reviewing and editing Incorporating input and
feedback Reusing and consolidating prior work efforts.

Features that fix Processes Lotus' team computing features and
functions address these processes with the goal to improve team
productivity and efficiency while lessening common team frustrations.
SmartSuite's team computing technology:  Eliminates the work involved
in tracking versions, Organizes the review process, Simplifies the
consolidation process, Manages distribution of and access to the work,
Creates common starting points, and Facilitates integration.

Lotus Word Pro 96, the word processor included in the upcoming
SmartSuite 96 Edition for OS/2 Warp, is the first word processor that
incorporates Document Versioning technology.  This feature enables
users to store multiple versions of a document in a sing le file.  No
longer do users need to devise clever naming conventions to manage
different document revisions.  Word Pro automatically preserves past
versions by storing only the changes made between versions.  Any
version can be used at any time.  Users ca n decide when to create a
new version -- either each time he opens the file or each time a new
person edits the file.
 
SmartSuite's TeamReview organizes the document review process with a
step-by-step guide.  This easy-to-learn feature lets an author
determine who should review the work and set the editing rights for
each reviewer.  The editing levels can range from "read o nly" to full
editing rights.  The author can also send individualized instructions
or greetings along with the document he wants reviewed.  It also does
not matter how the author distributes the document to the reviewers.
He has the choice of e-mail, Lotus Notes, floppy disk or network
distribution.  When the reviewers receive the document, they can use
Multi-Editor Revision Marking.  This tool lets the reviewers insert
notes where they want or make changes to the text without overwriting
the original work.

More than any other team task, consolidating input is by far the most
complex and tedious.  SmartSuite eliminates the tedium and simplifies
the complexity by automating the task with TeamConsolidate.  This
powerful tool collects each reviewer's edits and p uts them on one
screen so the author can see all changes, suggestions and notes at
once.  With all the input in one place, the author simply accepts or
rejects the edits and incorporates the suggestions he wants as he moves
through the document.  This aut omation eliminates the time-consuming
and confusing task of shuffling through files or papers to ensure that
each reviewers' comments are taken into consideration.

TeamMail gives document owners more power than ever before to manage
distribution and access.  With TeamMail, an owner can send his document
to a group all at once or select a sequence of recipients.  More than
that, TeamMail informs the document owner ab out its progress through
recipients.  SmartSuite also lets users send documents from within the
application without launching e-mail.  If users do use e-mail, they can
send documents as attachments or text.  Generally, TeamMail provides
fast, flexible distribution and access to documents.

To provide a common starting point for work that begins as several
source documents but will ultimately be delivered as one, SmartSuite
provides SmartMasters, templates for a wide array of documents.
Although earlier releases of SmartSuite have included SmartMasters,
Lotus has added functionality that supports Team Computing.  The
SmartMasters in SmartSuite '96 Edition for OS/2 Warp will include
TeamReview process information, such as editing rights.

On the other hand, if a team needs to combine output from different
applications, SmartSuite allows sharing of data and skills between its
applications.  Furthermore, SmartSuite automatically recognizes
documents not created with Word Pro.  Users can impor t these
documents, work with the document in Word Pro and then save it in the
format in which it came.  This lets SmartSuite users share work with
others even if they did not use the same tools.

Reaping Team Computing benefits If the trio of writers who created the
white paper had team computing tools at their disposal, their workload
would have consisted of developing content, agreeing to the flow of
ideas, and incorporating the best suggestions and input.  Document
Versioning would have simplified their work from the beginning.  They
would not have had to maintain two files for the different flows they
were considering.

If the team had used TeamReview to initiate the review process, they
could have given instructions to the reviewers to make their comments
on the soft copy version using multi-editor revision marking.
Furthermore with the flexibility of the revision marki ng, the
reviewers would have the flexibility to suggest changes and to make any
additional comments right to the document.  Even if one of the
reviewers wanted to do his edits from home, he would have no trouble
bringing home a floppy disk containing the document.  Overall, the
reviewers would have been so comfortable making their suggestions right
on-line that the team would not have had to bother to tell them to do
it on-line; they just would have wanted to do it that way.

The team also could have used the multi-editor revision marking to
avoid doing their initial editing work on hard copy.  This would have
eliminated several cumbersome steps:  faxing hard copies between the
two sites writing the comments on the hard copy transcribing the
comments from the hard copy to the soft copy

TeamConsolidate would have greatly simplified incorporating the
comments and suggestions from the reviewers.  Instead of having one
person go off and do the work manually and then show the others what he
had done.  They could have reviewed all the comments together on-line,
determining as they went along, what to keep, what to change, and what
to ignore.

Until Lotus made the leap to Team Computing technology, teams spent as
much or more time managing the logistics and mechanics of the team work
than they did doing the tasks germane to delivering their ultimate
goal.  Team Computing makes the hurdles associ ated with working as a
team negligible.  With SmartSuite '96 Edition, businesses can reap the
benefits of team work without having to endure the inefficiencies and
frustrations often associated with group projects.

Team Computing is coming soon to the OS/2 Warp platform.  Recently
introduced on the Windows platform, the SmartSuite 96 Edition for OS/2
Warp will follow shortly.  Stay tuned to this newsletter or to the
Desktop Products Section of the Lotus Web Site (ww w.Lotus.com) for the
latest announcements and information on availability for Lotus OS/2
applications.  Lotus expects to begin customer beta testing for Word
Pro 96 Edition for OS/2 Warp shortly.  Getting Started with OS/2 Warp

Staff Biographies

Jose De Jesus works as a contractor for IBM, debugging VDM and (soon)
WorkPlace Shell applications.  He is the author of "Borland Pascal With
Objects 7.0," published by MIS:Press in 1993.  He can be reached at
73141,317@compuserve.com.

Jeri Dube has been working for IBM for 12 years now.  She is currently
working as a Senior Marketing Support Rep, putting together
communication vehicles for OS/2 Warp and LAN Server.  She swears this
is the best job she has had in a long time.  Outside of I BM, Jeri
spends time with her family, scuba diving, and practicing Yoga.  Her
favorite quote is:  " Humanity has advanced, when it has advanced not
by being sober, cautious and responsible but by being playful,
rebellious and immature."Tom Robbins, Still Li fe with Woodpecker.
Contact Jeri at Internet:jdube@vnet.ibm.com.

Richard Hawes has worked in the PC industry for too long with a slight
detour into IBM mid-range systems when he joined IBM nearly seven years
ago.  Richard is now working in Basingstoke, England as a LAN Technical
Consultant in a European Project Office w here he provides assistance
to IBM support organisations.  Richard's spare time mainly revolves
around keeping his young daughter from wrecking the house.  His other
hobbies include swimming, cinema and juggling.  Low ceilings in his
house prevent him from j uggling five balls; otherwise, he would have
surely mastered it by now.  He can be reached at
Internet:juggler.vnet.ibm.com.

Steve McNally develops promotions with Marketing Reps as the Editor for
the IBM US Promotions and Price Actions Board.  He spends his free time
with his wife Kim and ten-month-old daughter, Olivia Hope, and tries to
put in as much time as possible writing fiction of all shapes and
sizes.  A quote that came to mind while composing his bio was "It's
like a sauna in here" -- Cosmo Kramer, in the sauna at a health club.
He can be reached at smcnally@vnet.ibm.com

Karla Myers is one of the original support technicians in the OS/2 Warp
Connect group in the Austin IBM center.  While her primary focus is
technical support for OS/2, she has also written "how-to" faxes for
customers and contributed to the latest revisio n of the Common
Questions that is shipped with OS/2 Warp Connect.  When she's not doing
customer support, she spends most of her spare time raising and riding
registered Appaloosas and Quarterhorses.  She can be reached through
the Internet at v$imyers.ausn otes@aussmtp.austin.ibm.com.

Collin Summers' company has been contracted by IBM to help in central
Illinois with OS/2 LAN Server and all those other PC Software type
stuff.  Before IBM took up all his time, he did independent computer
consulting in many areas, but focused on OS/2 wor k (Guess that's how
he ended up working with IBM).  Collin spends his free time, when there
is some, tinkering around and looking for new and better ways to use
computers.  Of course, he only runs OS/2 and OS/2 LAN Server on his
business' LAN.  Collin is av ailable at Internet:  summers@vnet.ibm.com
or 73422,3364@compuserve.com.



FEEDBACK Survey

We want to produce a quality newsletter that meets your needs.  Please
take a few minutes to answer the following questions and send your
responses to:  - FAX:  407-982-8888 - INTERNET ID:  JDUBE @
VNET.IBM.COM - MAIL:  Jeri Dube IBM Personal Software Products 1000
51st Street Boca Raton, Fl 334 - IBM VNET:  BCRVM1(WARPWIRE)

1. What other information would you like to see published in this
newsletter?
________________________________________________________________
_________________________________________________________________

2. Is the length of the newsletter:_____ Too short _____ Too long _____
Just right

3. Does this newsletter improve your product knowledge of OS/2 Warp and
LAN Systems products?  	_____ Yes _____ No

4. Do you currently have an OS/2 Warp or LAN systems product installed?
		_____ Yes _____ No

5. Do you plan to have an OS/2 Warp or LAN systems product installed?
		_____ Yes _____ No

6. Optional Name:  	 Address:  	 City, State, ZIP:  	 E-Mail Address:
		 COMMENTS:  Please offer any ideas/suggestions to improve the
newsletter.
_________________________________________________________________
_________________________________________________________________

Thank you for answering this survey.  We make improvements based on
what you tell us!

 OS/2 Warp and LAN Server Customer Reference Form If you would like to
share your experiences with OS/2 Warp or LAN Server, please fill out
the following form and get it back to us.  We are especially interested
in entrepreneurs who use OS/2 Warp or LAN Server and people who use
these products in unusual circumstances.

- FAX:  407-982-8888 - INTERNET ID:  JDUBE @ VNET.IBM.COM - MAIL:  Jeri
Dube IBM Personal Software Products 1000 51st Street Boca Raton, Fl
33434

GENERAL INFORMATION:  Name:  	 City:  	 State/Province:  	 Country:  	
Fax:  	 Email Address:  	 Street/P.O.Box:  	 City:  	 State/Province:
	 ZIP/Postal Code:  	

INDUSTRY:  ____Communications ____Cross Industry ____Distribution
____Education ____Finance ____Government ____Health ____Insurance
____Manufacturing ____Petroleum ____Process ____Transportation
____Travel ____Utilities

SIZE:  Annual Revenue US$:  ____< $5M ____$5 - 50M ____$51 - 100M
____$101 - 500M ____$501 - 999M ____$1 - 10B ____> $10B No.  of
Employees:  ____Small <100 ____Medium 100-500 ____Large > 500

No.  of Students if College/University:  ____<5000 ____5000 - 10000
____>10000

BACKGROUND:  (Briefly describe the business or situation where you used
the products.)
________________________________________________________________________
________________________________________________________________________
_____________________
___________________________________________________
________________________________________________________________________
________________________________________________________________________
_________________________________________________________
_______________
________________________________________________________________________
________________________________________________________________________

NEED:  (What problem/situation existed which caused you to seek a
solution using OS/2 Warp and/or LAN Server?)
________________________________________________________________________
______________________________________________________________________
__
________________________________________________________________________
________________________________________________________________________

DESCRIPTION OF SOLUTION:  (Please provide a detailed description.
Describe any mission critical applications.  For a networking solution,
include a description of the network.)
________________________________________________________________________
____
____________________________________________________________________
________________________________________________________________________
________________________________________________________________________
_______________________________________
_________________________________
________________________________________________________________________

BENEFITS:  (Please describe business or personal benefits.)
________________________________________________________________________
________________________________________________________________________
_________________________________________________
_______________________
________________________________________________________________________


OS/2 Warp MONTHLY NEWSLETTER
http://www.austin.ibm.com/pspinfo/newsletters.html Volume 9:  February,
1996

USER POWER WARP'D ETC.

