v11 Update?

[CDonahue]

Hi Aparajita,

 

It's been a little while since the last v11 update ... I was wondering
if you could apprise us of the latest status?

 

Thanks!

- Clayton

 

Clayton L. Donahue

The Healthcare Management Council, Inc.

cdonahue@... <mailto:cdonahue@...>

[phone]

 

_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[aparajita]

> It's been a little while since the last v11 update ... I was wondering
> if you could apprise us of the latest status?

The good: I'm killing myself to get this done.

The bad:  I'm killing myself to get this done, but the end is still  
weeks away.

The ugly: I have to check every line of code, and >50% of them need  
fixing or total rewriting. I have been battling with several open  
source libraries I am using (ICU, boost, mcrypt, logging) to get them  
compiled and working correctly. I'm running into bugs in the 4D plugin  
API. I have gone down several blind alleys and lost many days as a  
result. I went through 3 logging libraries before finding one that was  
usable.

What's actually done:

- Have a basic native Unicode plugin running on v11 on Mac OS X and  
Windows.

- There is now extensive logging of what's going on internally (with  
an option to turn debug logging on) so you and I know why things go  
wrong when they do.

- Using Unicode throughout.

- Lots of framework code rewritten.

- Have key file encryption/decryption working.

Doesn't seem like much, but the entire plugin rests on the framework,  
so getting that redone was essential. Now it's just a matter of  
ploughing through the plugin itself. I'm starting from plugin  
initialization and working my way through. At the moment I'm working  
on key file checking, and now that most of the foundational stuff is  
done I am moving much more quickly.

I have thrown out all planned enhancements except for ones that come  
with little or no effort. I have already sunk 100+ hours into this  
rewrite and it will require maybe 100+ more. I'm sorry to say I don't  
think I will ever recoup my investment in getting this version done,  
even with it being partially funded by a client. Initially it will  
probably be v11 only, with a 4D 2004 version coming later (if at all,  
depending on demand).

Regards,

    Aparajita
    www.aparajitaworld.com

    "If you dare to fail, you are bound to succeed."
    - Sri Chinmoy   |   www.srichinmoy.org


_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Walt Nelson-4]

Hello all.

In reading Aparajita's response regarding his v11 update to Active4d (see
below), it occurred to me that there is a lesson that we all can learn here.
-------------------

1) It's pretty clear that Vista is a failure; even Microsoft now knows that.
Hardware vendors are shipping their new Core Duo machines with Vista, but
with an accompanying DVD that allows the customer to downgrade to Windows
XP. After playing with Vista for a few days, I have made the decision to
downgrade - even though it requires a complete reformatting of the hard
drive and a complete re-install. The thing that pushed me over the edge is
the fact that I have a 4D program that checks every few minutes to make sure
that my 24/7 4D Webserver is running, and automatically re-launches 4D if it
has crashed. Now, with the new Vista security feature that requires the user
to click the OK button in another dialog when launching a program, this
means I will have to get under the hood and figure out how to OK that dialog
whenever 4D automatically re-launches, or my web server will stay down if it
crashes. Not good.

I believe that Vista failed as an upgrade not because of bugs, but because
of changes in behavior, like this one, that caused millions of programs and
procedures to break. Microsoft made a Big, Big, Big mistake.

2) v11: I my humble opinion, 4D Inc. made a similar mistake. So many changes
have been made in 4D behavior in v11, that virtually every major program
running in 4D will break when it is upgraded to v11. It will break not just
in one place, but in many, many, many places. I think that 4D is on the edge
of the cost-benefit curve: the cost of adjusting to v11 may become larger,
for many users, than the benefits to be gained. If this happens, they may be
surprised that customers will start demanding that a 4D 2004 license come
with every purchase of v11 - or the customers may move on to other products.
-------
What's done is done for 4D, but we, as developers, can learn from this: BE
CAREFUL ABOUT CHANGING UNDERLYING BEHAVIORS THAT AFFECT THE CUSTOMER. We
think that those changes are cool, but the customer sees them, in most
cases, as inconveniences. I am mostly developing for my own business, so I
can control this to some extent, but I still have one major client for whom
I develop about 4-6 hours a day. I'm working on a new module to be added to
our system, and since I am on retainer, I have the luxury of being able to
do "interactive enhancements," in which I make changes and get the
customer's feedback before those changes are carved in stone. Many times,
they will REJECT behavior changes because of the tax that those changes
create on user comfort levels. In other words, they want improvements, but
when it comes to basic behaviors that users have become accustomed to, they
want those behaviors to remain unchanged.
---------
After more than 20 years as a 4D consultant, this is a major revelation for
me. I think back on many of the problems that I have had with enhancements
and rewrites, and I can see this pattern occurring again and again. I am
amazed that I never saw it so clearly before.

So, for those of you who are still doing lots of development for lots of
clients, I suggest that you pay attention to this: be very careful when you
change basic program behaviors. Those changes can become major
customer-relationship-killers.

HTH,

Walt Nelson - Guam


-----Original Message-----
From: active4d-dev-bounces@...
[mailto:active4d-dev-bounces@...] On Behalf Of Aparajita
Fishman
Sent: Thursday, February 19, 2009 9:49 AM
To: active4d-dev@...
Subject: Re: [Active4d-dev] v11 Update?

> It's been a little while since the last v11 update ... I was wondering
> if you could apprise us of the latest status?

The good: I'm killing myself to get this done.

The bad:  I'm killing myself to get this done, but the end is still  
weeks away.

The ugly: I have to check every line of code, and >50% of them need  
fixing or total rewriting. I have been battling with several open  
source libraries I am using (ICU, boost, mcrypt, logging) to get them  
compiled and working correctly. I'm running into bugs in the 4D plugin  
API. I have gone down several blind alleys and lost many days as a  
result. I went through 3 logging libraries before finding one that was  
usable.

What's actually done:

- Have a basic native Unicode plugin running on v11 on Mac OS X and  
Windows.

- There is now extensive logging of what's going on internally (with  
an option to turn debug logging on) so you and I know why things go  
wrong when they do.

- Using Unicode throughout.

- Lots of framework code rewritten.

- Have key file encryption/decryption working.

Doesn't seem like much, but the entire plugin rests on the framework,  
so getting that redone was essential. Now it's just a matter of  
ploughing through the plugin itself. I'm starting from plugin  
initialization and working my way through. At the moment I'm working  
on key file checking, and now that most of the foundational stuff is  
done I am moving much more quickly.

I have thrown out all planned enhancements except for ones that come  
with little or no effort. I have already sunk 100+ hours into this  
rewrite and it will require maybe 100+ more. I'm sorry to say I don't  
think I will ever recoup my investment in getting this version done,  
even with it being partially funded by a client. Initially it will  
probably be v11 only, with a 4D 2004 version coming later (if at all,  
depending on demand).

Regards,

    Aparajita
    www.aparajitaworld.com



_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Peter Jakobsson-2]

On 19 Feb 2009, at 02:00, Walt Nelson wrote:

> virtually every major program
> running in 4D will break when it is upgraded to v11

Hi Walt

 From what I've seen of v11 so far, it looks like less pain to  
upgrade than, say, the move from 6.7 to 2003/4 since those had a  
brand new OS to support if you were using Mac.

Once the dust settles on v11 I'm firmly expecting to have a  
reasonably smooth upgrade for my 90-workstation 24/7 mission critical  
network apart from having to re-code a few sub-files which have been  
in there since the dawn of man. Also, Active4D of course.

[Will let you know if things don't go according to expectations -:) ]

Just wondering what in particular you would have had 4D do  
differently ? They are usually pretty good at preserving behavior via  
the compatibility checkbox approach. They had to support unicode to  
not go out of business. Same for SQL and X-LIFF. The native DB  
language (and behavior) is preserved. Legacy support for Ascii and  
resources is preserved.

Feel free to consider my comments naive as I haven't embarked on a  
single upgrade yet, so your the one with the experience on that front  
and I'll gladly take any advice going.

As for Active 4D, I've so far been provided with a great product for  
not much money by a 1-man operation so I don't feel in a position to  
make any demands there.  I've told my customer Mar-Dec 2010 for  
Active 4D v11 and I'm amazed at the progress Aparajita has already  
made according to his last report considering he was looking for work  
just a few weeks ago.

Think we'll just sit it out with 2004 for now and see how things go.

Regards

Peter

_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Walt Nelson-4]

Peter,

I haven't done any upgrades yet either, but I have read every message on this forum, and engaged in quite a few private discussions. And that's the reason I haven't done any upgrades yet. I don't have to put my hand in the fire if everyone who has tried it is saying that the fire is hot.

Go back and read the traffic on this list. My take is that the only ones who have gone live with their v11 apps and not suffered major pain are those who started over from scratch, like Bill, my old friend from Guam days. Only one developer with a converted app responded positively to the appeal for real-world v11 experiences, and he qualified his response by saying that his Foundation app, on the other hand, is giving him major headaches.

What would I like 4D to have done differently? GRADUALLY introduced all those new features, over a 3-5 year period. Spend more time and resources on compatibility issues, recognizing that customers don't have unlimited time and money to flush out bugs and strange behaviors. Don't change the way things work for everyone, to fix a problem that may have been experienced by a very few.

The reason I am speaking out is because no one else is. Everyone who is trying to use v11 is so focused on putting out individual fires, that they don't see that the entire forest is smoldering. Back in the "old days," there were many developers who would publicly hold 4D Inc. Accountable, but apparently those days are long gone.

It's a shame, though, because 4D as a product benefitted from those lively discussions. Changes were made because of public complaints, and we all benefitted. When I was in management at ACI US (now 4D Inc), we always knew that some customers were going to "hold our feet to the fire" when we made bad decisions or put out buggy versions.

Now, evertone seems afraid to say publicy what some are saying to me privately: "this upgrade has been a disaster."

If no one tells 4D that something is wrong, thsy are justified in thinking that everything is OK. So if things are not OK, it is our DUTY and our OBLIGATION to tell them - publicly and privately.

And that's MY $0.02.

Walt
Sent from my BlackBerry® wireless device

-----Original Message-----
From: Peter Jakobsson <lists@...>

Date: Thu, 19 Feb 2009 04:21:14
To: <active4d-dev@...>
Subject: Re: [Active4d-dev] VISTA and v11: A CAUTIONARY TALE



On 19 Feb 2009, at 02:00, Walt Nelson wrote:

> virtually every major program
> running in 4D will break when it is upgraded to v11

Hi Walt

 From what I've seen of v11 so far, it looks like less pain to  
upgrade than, say, the move from 6.7 to 2003/4 since those had a  
brand new OS to support if you were using Mac.

Once the dust settles on v11 I'm firmly expecting to have a  
reasonably smooth upgrade for my 90-workstation 24/7 mission critical  
network apart from having to re-code a few sub-files which have been  
in there since the dawn of man. Also, Active4D of course.

[Will let you know if things don't go according to expectations -:) ]

Just wondering what in particular you would have had 4D do  
differently ? They are usually pretty good at preserving behavior via  
the compatibility checkbox approach. They had to support unicode to  
not go out of business. Same for SQL and X-LIFF. The native DB  
language (and behavior) is preserved. Legacy support for Ascii and  
resources is preserved.

Feel free to consider my comments naive as I haven't embarked on a  
single upgrade yet, so your the one with the experience on that front  
and I'll gladly take any advice going.

As for Active 4D, I've so far been provided with a great product for  
not much money by a 1-man operation so I don't feel in a position to  
make any demands there.  I've told my customer Mar-Dec 2010 for  
Active 4D v11 and I'm amazed at the progress Aparajita has already  
made according to his last report considering he was looking for work  
just a few weeks ago.

Think we'll just sit it out with 2004 for now and see how things go.

Regards

Peter

_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/
_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Peter Jakobsson-2]

On 19 Feb 2009, at 11:09, Walt Nelson wrote:

> Everyone who is trying to use v11 is so focused on putting out  
> individual fires, that they don't see that the entire forest is  
> smoldering

Hey, you sure have conjured up some vivid imagery there. Certainly  
captured my imagination.

> Now, evertone seems afraid to say publicy what some are saying to  
> me privately:

That's 'cos they know they'll get batted into 'business list  
oblivion' on the official forum. Not a problem in itself but before  
4D took over the list hosting the old business list used to be quite  
lively. You got the feeling that there were a lot more subscribers (I  
think it was just called the 'public' list then - can't quite  
remember). I don't know why nobody wants to post on the biz list  
these days. They seem to think that nobody's listening so it results  
in 'rant bombs' getting periodically dropped on the tech list causing  
antagonism to mods & subscribers who want that list kept 'clean'.

Now we're hijacking Aparajita's list to discuss 4D business topics -  
how much more off-topic can you get than that :) ?, albeit I agree  
with you that these debates are healthy and necessary.

On 19 Feb 2009, at 11:09, Walt Nelson wrote:
> this upgrade has been a disaster

Wow, do you really think so ? I also have read all the list responses  
and didn't get this impression. Mine was that the problems were in  
proportion to the benefits and there have been many reports of  
performance gains, approval for industry standards support etc etc.  
v3.0 was pretty bad - you couldn't reliably run the new 4D server  
till about 3.2x but it was a technology leap that worked in the end.  
v6 wasn't rock solid till v6.0.6 r9 (I even remember the exact point  
revision that worked :)). 6.8 was so shaky I avoided it completely.  
2004.5 has been rock solid for us, so that's the background.  
Microsoft development products are even worse for quality control in  
my experience.

I've got a huge network to upgrade presently, I'd like to think I was  
delivering solutions (mainly more scaleability) not 'new' problems.  
They've already asked me for quotes for licensing and my own time. I  
don't really want to quote a fixed price for time. Major risk factor  
given we're heading into the unknown.


> What would I like 4D to have done differently? GRADUALLY introduced  
> all those new features

Yes - I agree this would have been the alternative. I certainly don't  
need all the new stuff and I'm not going to be using SQL, but I  
suppose that it's a difficult call because people also give 4D a hard  
time for charging upgrade prices for minimal new functionality.

You wouldn't get much argument from developers across the board that  
reliability is about top of everyone's list of priorities. It's a  
complex product and only getting more complex so from that point of  
view I'd be happy to take less functionality for more reliability. I  
also would very much underwrite your call for more open discussion of  
business issues. That problem can't all be blamed on 4D for taking  
over the list hosting, there are plenty of places where people can  
post but they don't. e.g. the biz list on iNUG or the google groups  
list which is active from time to time.

In the meantime, thanks to Aparajita for indulging us on here.

Peter

_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Peter Jakobsson-2]

On 19 Feb 2009, at 02:00, Walt Nelson wrote:
> *several notable remarks*


I suppose that, following the thread just opened over on the iNUG,  
you rest your case :)

It still will be a pretty remarkable product if it can be got stable  
though. Also, Active4D remains unique in it's class as far as making  
4D solutions a viable proposal for taking applications online. I hope  
it can all get through the rocky phase.

Peter

_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/

[Mehboob Alam]

> The good: I'm killing myself to get this done.

Thanks for including the original message from Aparajita, that didn't even show up in my inbox, so I'm mystified..

sincerely,
mehboob alam


Electricity is actually made up of extremely tiny particles called electrons, that you cannot see with the naked eye unless you have been drinking. - Dave Barry



     
_______________________________________________
Active4D-dev mailing list
Active4D-dev@...
http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev
Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/