PDA

View Full Version : Re: Download Update option



Stephen W. Meeley
1-Jun-2005, 09:35 AM
Chris (and others),

Do you have a list of specific bugs that you feel fall into the category
of "bug issues that effect the end-user applications"? I understand that
there are lots of additions or changes that you (and others) will feel
have tremendous end-user impact, and I'm not trying to ignore those
types of things at all.

I just want to start a specific conversation focused on known bugs that
you feel would "get them (them being end-users) some benefit".

Once I get a good handle on which bugs can give the end-user impact you
are looking for (and we do want to provide), we can move on to other
areas.

I will add one thing: keep in mind that many of the changes we make that
may seem at first blush to be more development oriented are really there
in support of the changes we (and developers) make (or want to make)
that will have tremendous end-user impact. The underlying development
system needs to be solid and critical technologies need to be supported
in order to have the building blocks necessary to take advantage of
what's out there. FlexCOM and Web Services (and then the addition of
struct and array support in both those areas) are great examples of
this. They don't give you a specific new control to use, they open up
entire new worlds to you (and us too).

In any case, let's take this "how to get end-user impact" discussion
forward, starting with bugs.

Best regards,

-SWM-


-----Original Message-----
From: Chris Spencer [mailto:nomail@nomail.com]
Posted At: Tuesday, May 31, 2005 6:14 PM
Posted To: product-direction
Conversation: Download Update option
Subject: Re: Download Update option


David

I would not like to see DAWs resources funnelled of to do this.
Updates don't come from DAW at such a pace that it is difficult to keep
up.
A simple note in the NG is all that is required.

However on the topic of updates I would prefer to see resources now
spent on runtime and bug issues that effect the end-user applications.
I am happy to work within the development environment as it is now.

DAW have in place a client subscription model and to make that
attractive (and saleable) to end-users the end-user needs to see that
paying there $$$s gets them some benefit. All the fixes and improvements
in the development environment amount to zero for the end-user. (There
can be the argument that better tools means better response to users but
at the end of the day the user does not see anything.)

Things that the client sees are like in the last release richEdit
classes etc. I would like to see the developmenyt in these areas. There
are many more data presentation tools and classes that could be provided
to make the end-users experience a good one

Chris Spencer
TUFware Systems

"David Martinko" <RedeemedSoftware@Hotmail.com> wrote in message
news:Xq8RSXeZFHA.1312@dacmail.dataaccess.com...
> I'ld like to see an option in the studio to check for updates for
> installed components and allow for optional download.
>
> I think it should resemble the product registration program... and
> list all updates available. A user should not be forced to upgrade to
> the latest (especially since the latest may still be a beta). Updates
> are for the current version of each product, so a registration for VDF

> 9 does not tell you and update is available because VDF 10 is out
> there. It should only notify if updates for VDF 9 are available.
>
> All that should be needed is a webservice. I would open my IDE, goto
> the Help menu and select Check for Updates. The IDE would generate an
> XML document of all VDF registrations on my PC (same thing that
> appears in the product registration). This is transmitted to DAC via a

> webservice. The returned XML document is modified to list all product
> updates available for download, including the path to each download. A

> list of these updates are displayed to the user who can then decide to

> install all updates, a few, or none.
>
> If possible, the list might also display which updates are a part of
> the developer's subscription model and which aren't (using the serial
> number, I imagine).
>
> Since this would be a program launched from the IDE, and not a part of

> the IDE, it could be redistributable to end-users to make sure their
> RC is up-to-date. If a client's machine wasn't working properly with
> the latest install, I could have then run the Check for Updates
> program to make sure they have the latest client runtime.
>
> Any thoughts?
>
> --
> David Martinko
> Redeemed Software
> 248-535-7495
> RedeemedSoftware(SHIFT+2)Hotmail(PERIOD)com
> www.redeemedsoftware.com
>
>
>

Chris Spencer
1-Jun-2005, 06:26 PM
Stephen

First thanks for wanting to start a specific thread on this topic
I will get back with more when I have a chance but a few straight away are:

1. Manifest Files
When using Manifest files the combos, spinforms, buttons take on the XP
style but the forms do not giving a strange look. I personally do not use
manifest files, preferring at least an overall simolar chiselled look in
preference to the look provided by using the manifest files.

2. FlexCom
Even with all the advances in Coms handling there are still issues with
memory leaks

3.GUI Controls
Where is the rebar control promised way back?
Where is the rewritten Grids that were not subject to number of element
issues?

Chris Spencer
TUFware Systems



"Stephen Meeley" <stephen-m@dataaccess.com> wrote in message
news:9fc0ycrZFHA.4508@dacmail.dataaccess.com...
Chris (and others),

Do you have a list of specific bugs that you feel fall into the category
of "bug issues that effect the end-user applications"? I understand that
there are lots of additions or changes that you (and others) will feel
have tremendous end-user impact, and I'm not trying to ignore those
types of things at all.

I just want to start a specific conversation focused on known bugs that
you feel would "get them (them being end-users) some benefit".

Once I get a good handle on which bugs can give the end-user impact you
are looking for (and we do want to provide), we can move on to other
areas.

I will add one thing: keep in mind that many of the changes we make that
may seem at first blush to be more development oriented are really there
in support of the changes we (and developers) make (or want to make)
that will have tremendous end-user impact. The underlying development
system needs to be solid and critical technologies need to be supported
in order to have the building blocks necessary to take advantage of
what's out there. FlexCOM and Web Services (and then the addition of
struct and array support in both those areas) are great examples of
this. They don't give you a specific new control to use, they open up
entire new worlds to you (and us too).

In any case, let's take this "how to get end-user impact" discussion
forward, starting with bugs.

Best regards,

-SWM-


-----Original Message-----
From: Chris Spencer [mailto:nomail@nomail.com]
Posted At: Tuesday, May 31, 2005 6:14 PM
Posted To: product-direction
Conversation: Download Update option
Subject: Re: Download Update option


David

I would not like to see DAWs resources funnelled of to do this.
Updates don't come from DAW at such a pace that it is difficult to keep
up.
A simple note in the NG is all that is required.

However on the topic of updates I would prefer to see resources now
spent on runtime and bug issues that effect the end-user applications.
I am happy to work within the development environment as it is now.

DAW have in place a client subscription model and to make that
attractive (and saleable) to end-users the end-user needs to see that
paying there $$$s gets them some benefit. All the fixes and improvements
in the development environment amount to zero for the end-user. (There
can be the argument that better tools means better response to users but
at the end of the day the user does not see anything.)

Things that the client sees are like in the last release richEdit
classes etc. I would like to see the developmenyt in these areas. There
are many more data presentation tools and classes that could be provided
to make the end-users experience a good one

Chris Spencer
TUFware Systems

"David Martinko" <RedeemedSoftware@Hotmail.com> wrote in message
news:Xq8RSXeZFHA.1312@dacmail.dataaccess.com...
> I'ld like to see an option in the studio to check for updates for
> installed components and allow for optional download.
>
> I think it should resemble the product registration program... and
> list all updates available. A user should not be forced to upgrade to
> the latest (especially since the latest may still be a beta). Updates
> are for the current version of each product, so a registration for VDF

> 9 does not tell you and update is available because VDF 10 is out
> there. It should only notify if updates for VDF 9 are available.
>
> All that should be needed is a webservice. I would open my IDE, goto
> the Help menu and select Check for Updates. The IDE would generate an
> XML document of all VDF registrations on my PC (same thing that
> appears in the product registration). This is transmitted to DAC via a

> webservice. The returned XML document is modified to list all product
> updates available for download, including the path to each download. A

> list of these updates are displayed to the user who can then decide to

> install all updates, a few, or none.
>
> If possible, the list might also display which updates are a part of
> the developer's subscription model and which aren't (using the serial
> number, I imagine).
>
> Since this would be a program launched from the IDE, and not a part of

> the IDE, it could be redistributable to end-users to make sure their
> RC is up-to-date. If a client's machine wasn't working properly with
> the latest install, I could have then run the Check for Updates
> program to make sure they have the latest client runtime.
>
> Any thoughts?
>
> --
> David Martinko
> Redeemed Software
> 248-535-7495
> RedeemedSoftware(SHIFT+2)Hotmail(PERIOD)com
> www.redeemedsoftware.com
>
>
>

wila
2-Jun-2005, 04:19 AM
SWM,

Stephen Meeley wrote:
> In any case, let's take this "how to get end-user impact" discussion
> forward, starting with bugs.

Not wanting to be disrespectful or anything like that, but don't you
have all serious bugs already logged into your system?

To quote you from an earlier reply to me:
<quote>
Do not confuse being logged (or not) with being fixed (or not) - those
are two separate states that are not (and should not be) mutually
exclusive.
</quote>

I for one don't like having to repeat myself again and again in order to
be heard. You have the bugs in your system. The ones that have end-user
impact can be deduced from the bug descriptions.

Personally i think it's pretty obvious that VDF is lacking features in
the presentation area for building nice user interfaces. Don't forget
that the UI is the first thing the end-user / prospect / customer sees.
Marketing has proven time and time again that having a nice looking UI
and a crappy application still sells way better as a lousy UI on a great
application. Of course doing both of them right the first time is
preferable.

But here's some examples:
1. Bitmaps are supported, but if you want to display a bitmap with more
as 256 colors it is considered to be too fancy for VDF and you need to
get an external control or work around it by displaying it for instance
in an Internet Explorer activex control (hello overhead), by using my
cvBitmapContainer or by buying a more capable control from Starzen (or
others)

2.a. RTF is supported as a native control nowadays (great!) but you
cannot print RTF using the native printing solution, rendering it a bit
funny as complete solution. So you need crystal, but not crystal XI as
that won't work. You can print the text unformatted, but you need a
workaround again and being able to only print unformatted text kind of
makes the rich text control look foolish.

2.b. RTF is supported as a native control, but you cannot save >16kb
fields using the embedded database system unless you work around the
recordsize. Ok.. ok.. you can (apparently) use BLOBS in other databases,
but i'm unsure which ones as i haven't found it documented (which could
be my fault)

3. UNICODE support, it has been on my list of request for years, can't
even remember how many times i brought that up.

4. fancy controls, fancy controls, fancy controls to quote Steve
Ballmer, o no he said developers, developers, developers :)
There's a number of standard windows widgets that are just not supported
or lack features. Look for example at how an application with the
cWindowsEx framework looks in comparison to a plain VDF application...

--
Wil

Chris Spencer
2-Jun-2005, 04:46 AM
Adding to Wils button section. In some other development languages buttons
have the ability to have Text and bitmaps and be positioned in 9 positions
(top left, top centre etc)

I agree 100% with Wil on
> Marketing has proven time and time again that having a nice looking UI and
> a crappy application still sells way better as a lousy UI on a great
> application. Of course doing both of them right the first time is
> preferable.


Chris Spencer
TUFware Systems

"Wil van Antwerpen" <info@antwise.com> wrote in message
news:%23DjetQ1ZFHA.3896@dacmail.dataaccess.com...
> SWM,
>
> Stephen Meeley wrote:
>> In any case, let's take this "how to get end-user impact" discussion
>> forward, starting with bugs.
>
> Not wanting to be disrespectful or anything like that, but don't you have
> all serious bugs already logged into your system?
>
> To quote you from an earlier reply to me:
> <quote>
> Do not confuse being logged (or not) with being fixed (or not) - those
> are two separate states that are not (and should not be) mutually
> exclusive.
> </quote>
>
> I for one don't like having to repeat myself again and again in order to
> be heard. You have the bugs in your system. The ones that have end-user
> impact can be deduced from the bug descriptions.
>
> Personally i think it's pretty obvious that VDF is lacking features in the
> presentation area for building nice user interfaces. Don't forget that the
> UI is the first thing the end-user / prospect / customer sees. Marketing
> has proven time and time again that having a nice looking UI and a crappy
> application still sells way better as a lousy UI on a great application.
> Of course doing both of them right the first time is preferable.
>
> But here's some examples:
> 1. Bitmaps are supported, but if you want to display a bitmap with more as
> 256 colors it is considered to be too fancy for VDF and you need to get an
> external control or work around it by displaying it for instance in an
> Internet Explorer activex control (hello overhead), by using my
> cvBitmapContainer or by buying a more capable control from Starzen (or
> others)
>
> 2.a. RTF is supported as a native control nowadays (great!) but you cannot
> print RTF using the native printing solution, rendering it a bit funny as
> complete solution. So you need crystal, but not crystal XI as that won't
> work. You can print the text unformatted, but you need a workaround again
> and being able to only print unformatted text kind of makes the rich text
> control look foolish.
>
> 2.b. RTF is supported as a native control, but you cannot save >16kb
> fields using the embedded database system unless you work around the
> recordsize. Ok.. ok.. you can (apparently) use BLOBS in other databases,
> but i'm unsure which ones as i haven't found it documented (which could be
> my fault)
>
> 3. UNICODE support, it has been on my list of request for years, can't
> even remember how many times i brought that up.
>
> 4. fancy controls, fancy controls, fancy controls to quote Steve Ballmer,
> o no he said developers, developers, developers :)
> There's a number of standard windows widgets that are just not supported
> or lack features. Look for example at how an application with the
> cWindowsEx framework looks in comparison to a plain VDF application...
>
> --
> Wil