Challenging the users in the MultiValue community to produce an Application that runs on a Thumb-Drive

On Feb 21, 3:58*pm, Ross Ferris <ro... (AT) stamina (DOT) com.au> wrote:

Unless there have been some developments that I'm not aware of, I
would have thought that rather than having their head in the sand,
these guys have said "you HAVE to use QM for this challenge", because
AFAIK that is the only MV product that is capable of running
standalone on the thumb drive (pretty sure all of the other players
would need to install registered components, though happy to be wrong)

Loading an applicaiton from a thumb drive is hardly rocket science.
The fact that other systems have not been deliberately packaged to do
this indicates only that:

1) Nobody is really asking to do it;
2) It is regarded as a trivial thing to do;

Needless to say with a system such as jBASE, which allows you to
separate the components you need form those you do not in a trivial
manner (as in include them or don't), it is rather pointless to create
a thumb drive 'version' as how do jBASE know what components should be
included? As no MV system is the operating system itself anymore, then
booting from a USB stick involves booting something that already
exists.

So in fact I find the whole 'challenge' to be completely meaningless
if it is meant to mislead people in to thinking that only OpenQM can
'do' this. It is more likely that they know any system can load from
the stick and so they do not limit the size of it so that people
running U2 could enter. Depending on what you are deploying, you only
need a few hundred KB for a jBASE app.

However, as all licensing models currently make you buy licenses
anyway, then the exercise is pretty pointless.

"Jim Idle" <jimi (AT) temporal-wave (DOT) com> wrote

On Feb 21, 3:58 pm, Ross Ferris <ro... (AT) stamina (DOT) com.au> wrote:
Unless there have been some developments that I'm not aware of, I
would have thought that rather than having their head in the sand,
these guys have said "you HAVE to use QM for this challenge", because
AFAIK that is the only MV product that is capable of running
standalone on the thumb drive (pretty sure all of the other players
would need to install registered components, though happy to be wrong)

Loading an applicaiton from a thumb drive is hardly rocket science.
The fact that other systems have not been deliberately packaged to do
this indicates only that:

1) Nobody is really asking to do it;
2) It is regarded as a trivial thing to do;

Needless to say with a system such as jBASE, which allows you to
separate the components you need form those you do not in a trivial
manner (as in include them or don't), it is rather pointless to create
a thumb drive 'version' as how do jBASE know what components should be
included? As no MV system is the operating system itself anymore, then
booting from a USB stick involves booting something that already
exists.

So in fact I find the whole 'challenge' to be completely meaningless
if it is meant to mislead people in to thinking that only OpenQM can
'do' this. It is more likely that they know any system can load from
the stick and so they do not limit the size of it so that people
running U2 could enter. Depending on what you are deploying, you only
need a few hundred KB for a jBASE app.

However, as all licensing models currently make you buy licenses
anyway, then the exercise is pretty pointless.

For Jim I mainly.

It wasn't completely wasted. Like the Japanese at Pearl Harbor it awakened a
sleeping giant.

BobJ

Jim,

Let me give you an example of how this is being used (not the only
one)....

QM is being used in educational environments. The students can carry
their entire project around on a USB stick and plug it into any
convenient Windows PC. Nothing needs to be installed on the PC at all.
In typical use, they take the USB stick to the college to work on
during the day and then work on it at home in the evening.

Loading an applicaiton from a thumb drive is hardly rocket science.

True but most of the other systems require software components to be
installed on the PC (e.g. services).

The fact that other systems have not been deliberately packaged
to do this indicates only that:
1) Nobody is really asking to do it;

Odd. We have a site with over 130 separate USB stick installations of
QM. I don't know details of how they use it but it was an essential
capability for them.

2) It is regarded as a trivial thing to do;

It was not trivial but neither was it excessively difficult.


If the original "challenge" was deliberately set as an "only QM can do
this" task, I must agree that it was not really fair.


Martin Phillips, Ladybridge Systems

Jim Idle wrote:

On Feb 21, 3:58 pm, Ross Ferris <ro... (AT) stamina (DOT) com.au> wrote:
Unless there have been some developments that I'm not aware of, I
would have thought that rather than having their head in the sand,
these guys have said "you HAVE to use QM for this challenge", because
AFAIK that is the only MV product that is capable of running
standalone on the thumb drive (pretty sure all of the other players
would need to install registered components, though happy to be wrong)

Loading an applicaiton from a thumb drive is hardly rocket science.
The fact that other systems have not been deliberately packaged to do
this indicates only that:

1) Nobody is really asking to do it;
2) It is regarded as a trivial thing to do;

Needless to say with a system such as jBASE, which allows you to
separate the components you need form those you do not in a trivial
manner (as in include them or don't), it is rather pointless to create
a thumb drive 'version' as how do jBASE know what components should be
included? As no MV system is the operating system itself anymore, then
booting from a USB stick involves booting something that already
exists.

So in fact I find the whole 'challenge' to be completely meaningless
if it is meant to mislead people in to thinking that only OpenQM can
'do' this. It is more likely that they know any system can load from
the stick and so they do not limit the size of it so that people
running U2 could enter. Depending on what you are deploying, you only
need a few hundred KB for a jBASE app.

However, as all licensing models currently make you buy licenses
anyway, then the exercise is pretty pointless.

I also don't see a significant point for the challenge. Nevertheless
it's worth considering for the simple discipline of conjuring / hacking
a response.

With a QM solution there is simply no challenge at all: no special
discipline by any app-builder as the tool-builder did all the work.
With a jBase app running on the host machine's os, is any host-resident
registering required? (There is no constraint to run on a windows
host.) There is no constraint on what else is on the thumb drive, such
as a portfolio of other apps to interact with. Specifically there is no
constraint on using the thumb drive as a bootable drive containing
whatever app + related apps + os + mv app server reside there. So the
effort to enter is open to all, with a back-challenge to the challenger
to clarify how to score a "win" besides "it works" or "it doesn't work".

The more important issues would be any commercial challenges. Under
what conditions need you transport only a thumb drive and not a
processor? I can only suppose those involving hazardous journeys in
which the cost and risks of carrying a processor in addition to the
memory are excessive. I cannot even imagine a mars mission offers such
a limitation. Perhaps your robotic transporter has the capacity for
only a single thumb drive? Perhaps the traveling consultant cannot heft
a netbook but needs not only to borrow the client's clock to tell her
what time it is, but her computer too?

Licensing questions are interesting. In the old days when my disk
failed I needed to transfer my license. The history has the licensing
to the disk / thumb drive, so it could be used just as if it were its
own computer too. Who knows how this might work, were thumb drive
applications important? What makes them important, however, in this
viewpoint is the low cost of distribution. As with the QM demo it is a
simple download or copy.

When combined with EasyCo's usb Supercharger for flash memory, the
functioning thumb drive mv app is as effective as if run on a spinning
disk drive.

The Challenge is much more of a business challenge than a technical
challenge. What type of an application is usefully deployed this way?
What type of deployment problem is solved by this licensing model?

-- Bill Cooke

I'm a little startled by some of the comments. On one side there's
the "what good is an app on USB" and on the other is "it's so easy a
caveman could do it".

To the first: many people move from one location to another and want
to take not only their data but the programs that operate on the data.
In many cases we find ourselves in an area where there are computers
but we don't have permission to load anything. Frankly I'm astounded
at just about any "why would anyone want this" question these days
when the world is clearly and constantly enamored with new gadgets and
new ways to do things.

To the second: Once again Ladybridge makes a splash because QM is the
first platform to support some feature, when the reality (pun
intended?) is that other platforms may be able to do the same, they
just don't tell anyone. While QM is truly innovative in some ways, I
think Martin has simply done better at telling people what his
software does, and getting them to appreciate it, than his
competitors.

Now I have no idea if jBASE can or cannot run code from USB. I
suspect it easily can because jBASIC compiles to OS-level executables,
so a jBASE EXE or DLL are no different than any other. The question
is what resources does jBASE need, what services are required for
locking or other functions? (Noting that with a "single user" USB
many functions aren't required.)

We simply don't know what the limits or potential of jBASE are for USB
because the current owner/developers, like most of their competitors,
simply don't tell us. In many cases the people who support the
platforms really don't have much insight or vision into how the
software can be used in the field. In other cases there is a
disconnect between Engineering and Marketing/Sales. I think this is
the case for most of the MV vendors. The result is that they can't
provide vision to this audience about the strengths of the platform,
so many end-users look elsewhere to satisfy their requirements.

Nuf outta me...

T

On Feb 19, 3:17*pm, "frosty" <fros... (AT) bogus (DOT) tld> wrote:

Haven't seen this mentioned yet here in cdp:

The Colorado MultiValue Users Group is Challenging
the users in the MultiValue community to produce an
Application that runs on a Thumb-Drive.

Even if we define the exact challenge:

1. Porting one of the Pick versions to a thumb drive

2. Writing an application that is compact enough to run on a thumb
drive

3. Both

any of those possibilities is not a challenge for Pick at all but it
may mean that we are getting some fresh blood on this list. All we
need to do now is encourage them to read the History of Pick first.

Henry Keultjes
Mansfield Ohio USA

Before I respond to Henry I'll toss these out:
http://en.wikipedia.org/wiki/Portable_software
Specific products:
http://en.wikipedia.org/wiki/VMware_ThinApp
http://en.wikipedia.org/wiki/Ceedo
http://en.wikipedia.org/wiki/RUNZ (Linux)
And it would be funny to run U2 on U3:
http://en.wikipedia.org/wiki/U3

hbkeultjes wrote:

Even if we define the exact challenge:
1. Porting one of the Pick versions to a thumb drive
2. Writing an application that is compact enough to run on a thumb
drive

Henry, with thumb/USB drive capacity now in the area of 256 GigaBytes,
the concept of "compact" is purely a reference to the physical size,
not data storage.

3. Both

any of those possibilities is not a challenge for Pick at all but it
may mean that we are getting some fresh blood on this list.

I don't think a thumb drive challenge would be something for people
who aren't familiar with the platform. This is where I'd say "what's
the point?" If there is a challenge intended to get new people to
learn the DBMS, don't limit the scope to hardware that only one or two
MV platforms can support.

All we need to do now is encourage them to read the
History of Pick first.

IMO that's exactly the way Not to start introducing people to the DBMS
- by talking about how old it is, radical roots, character screens,
etc. Anyone who is going to learn the platform should just learn how
to use it for modern use in comparison with RDBMS alternatives. The
fact that it has a rich history is just a testament to the viability
of the model.

T

I used D3/Linux.. ( Ancient 7.1 version )

Mounted a thumb-drive as a file system. Installed D3 on it. Declared
the D3 virtual disk as a Linux file .Did a restore of one of my ham
radio applications ( I know - not a commercial app ).

Kicked up D3, Logged into the app... and off it all went..Not
blazingly quick, but it works.

Boring really. What is all the fuss about ?

Gee

PS: Do I win ?



On 25 Feb, 18:08, Tony Gravagno
<address.is.in.po... (AT) removethis (DOT) com.invalid> wrote:

Before I respond to Henry I'll toss these out:http://en.wikipedia.org/wiki/Portable_software
Specific products:http://en.wikipedia.org/wiki/VMware_...iki/RUNZ(Linux)
And it would be funny to run U2 on U3:http://en.wikipedia.org/wiki/U3

hbkeultjes wrote:
Even if we define the exact challenge:
1. Porting one of the Pick versions to a thumb drive
2. Writing an application that is compact enough to run on a thumb
drive

Henry, with thumb/USB drive capacity now in the area of 256 GigaBytes,
the concept of "compact" is purely a reference to the physical size,
not data storage.

3. Both

any of those possibilities is not a challenge for Pick at all but it
may mean that we are getting some fresh blood on this list.

I don't think a thumb drive challenge would be something for people
who aren't familiar with the platform. *This is where I'd say "what's
the point?" *If there is a challenge intended to get new people to
learn the DBMS, don't limit the scope to hardware that only one or two
MV platforms can support. *

All we need to do now is encourage them to read the
History of Pick first.

IMO that's exactly the way Not to start introducing people to the DBMS
- by talking about how old it is, radical roots, character screens,
etc. *Anyone who is going to learn the platform should just learn how
to use it for modern use in comparison with RDBMS alternatives. *The
fact that it has a rich history is just a testament to the viability
of the model.

T

Gee wrote:

PS: Do I win ?

If I understand what you did correctly then I don't think so. You're
just using the device for storage. What happens if you move it to
another system where D3 isn't installed? Even if you installed D3 on
a different system and used the device for data, the activation
signature wouldn't match so the system would be unusable.

As I understand it the topic involves the ability to walk up to any
computer, plug in a thumb drive, execute an onboard application, then
take the app and data away with you without leaving any residue.

The software I referenced at wikipedia facilitates this - some may
work for MV.

T

Hi T

Good point... Back to the drawing board.

G





On 26 Feb, 20:48, Tony Gravagno
<address.is.in.po... (AT) removethis (DOT) com.invalid> wrote:

Gee wrote:
PS: Do I win ?

If I understand what you did correctly then I don't think so. *You're
just using the device for storage. *What happens if you move it to
another system where D3 isn't installed? *Even if you installed D3 on
a different system and used the device for data, the activation
signature wouldn't match so the system would be unusable.

As I understand it the topic involves the ability to walk up to any
computer, plug in a thumb drive, execute an onboard application, then
take the app and data away with you without leaving any residue.

The software I referenced at wikipedia facilitates this - some may
work for MV.

T