Arrrrrrrrrrggggggggggghhhhhhhh

[Progress_Guy]

Dear god give me strength - Although I suspect he has more pressing matters at the moment !!!

As a die hard Progress Developer I've always stood by the Progress App. Builder, but, its advantages are beginning to get fewer and fewer, i.e. the only advantage I can now see is :-

1. The ability to seamlessly integrate between the code (using Progerss 4GL) and the database.

Here endeth the advantages.

Now disadvanteges :-

1. Has this nasty habbit of corrupting an application whilst developing it - examples of this are too numerous to mention, but, I've just spent a morning rewriting a window because it somehow allowed me to resize the window and promptly resize and position all widgets and frame and so - nice !! Other examples include corrupt Smart Object procedures and an annoying message which pops and then if you save promptly corrupts all code. So mad I cannot 'member what it is off the top of my head.
2. Extremely 'twee' and old fashioned interface.
3. Colours - come on Progress I don't want to have to modify a config file to distribute these.
4. Its concept of MDI (Multiple Document Interface) is limited to say the least.
5. Some sort of repository for holding details about procesures so that when you type in a procedure name the parameters you need are automatically populated. This alone would save masses of development time, particularly with App. Server. - yeah, I know that .net use this, but, come on progress a little plagiarism wouldn't go amiss from time to time.

To top this I'm doing my MCAD and dot net interfaces are just so superior. Progress App. Builder now looks in excess of 10 years old.

I was even more dismayed upon glancing at OE 10 that it looks about the same !!!!

I suppose I could go to Progress and whinge, but, all that Progress are interested in these days is Dynamics - Yeah and how many are using this (in the UK not many - wonder why ?? ). I started off using ADM1 and the chances of me ever using a standard Progerss framework again are negative a zillion.

And if I replace all our front-ends with .net then what would my superiors be saying - well why don't we look at alternatives for the databases - say SQL Serevr or Oracle. Good questions - I may just take them up on that. At least it'd be easier to get a new job.

Come on Progress guys are you just giving up on it and letting everyone go the .net front-end !!

Can you tell I'm a little peeved. Maybe I should have had a brew and counted to 10 before writing this.

I need a lie down now.

 

[Lee Curzon]

Dave,

I don't have time for a long post, but my understanding of Progress current direction is there is a recognition that it can't hope to compete with Microsoft and associated 3rd Party vendors on the front end - er- front.

By strengthening its back-end services, and enhancing its integration facilities, it can stay alive in a market otherwise dominated by a couple of big players.

At Progress core is its 4GL, which marks it out from the rest. Using this to provide business services, and looking to (say) Microsoft for the front end seems the way to go for people like me. I will miss ADM2 though...

Also remember that Eclipse will be taking over from Appbuilder in the long run.

Lee

 

[Progress_Guy]

Thanks Lee - I was having a womans version of a 'bad hair day' and losing a considerable amount of work made me quite cross.

To put a calmer perspective on things I do love the integration between the 4GL and the App. Builder, but, it is sooooo frustrating at times.

Cheers,
Dave.

P.S. Excuse the typos on last posting. I was shaking with temper.

 

[Lee Curzon]

Thanks Lee - I was having a womans version of a 'bad hair day'

You suprise me. With a name like 'Dave' I assumed you were a man

 

[Lee Curzon]

I've posed this question a couple of times on related threads:

Is there anything in the Microsoft arena that is comparable to ADM2 Smart Objects?

If there is, is there anyway it can be integrated (non ODBC) with Progress backend?

 

[Lee Curzon]

losing a considerable amount of work made me quite cross.

Being rather anal about this sort of stuff, I have an hourly, twice-daily and daily backup of my work-in-progress running in the background (in addition to the manual copies I make for each bug fix).

I use the wonderful freeware Syncback:

http://www.2brightsparks.com/downloads.html


Always useful for those 'Appbuilder moments'.

 

[joey.jeremiah]

You can add a small script to the adecomm/_adeevnt.p ( ADE integration hook ) to save back N number of versions whenever you save anywhere in the ADE editor/appbuilder

I also use it to separate source code and r-code files into gui and src directories whenever the appbuilder compiles

You can find references in the "Basic Development Tools" doc
Tell me if you need a hand ...

A few years back when I was a junior developer at some shop, the company's owner "forgot" ( I think it slipped his mind for a... decade ) to pay his taxes and one day a group of 30 IRS agents stormed the offices and literally ripped the server off the wall. Now on top of that the dumb *** incharge of the IT department didn't have a complete backup scheme inplace and although the database did have a backup some work and documents didn't and a few years of work were lost.

Ever since this life changing experience I'm crazy about backups.
It wasn't all that bad he did get a nice picture in the news

As Lee Curzon mentioned there have been a great deal of changes in the ADE with the move to Eclipse in OE10.1A including the addition of the BIRT report builder

You can find references and screen captures on the new ADE in the exchange 2005 presentations, apparently it got a standing ovation


But I don't entirely disagree with you

ON WITH THE REVOLUTION AND OFF WITH THEIR HEAD !!!!

 

[mpowell_esq]

RE
Is there anything in the Microsoft arena that is comparable to ADM2 Smart Objects?

Smart Objects & ADM / ADM2, is a (PSC / PRGS) Progress thing. Does Progress have anything comparable to the pile of brown stuff that is Windows? Answer NO. You are comparing apples with oranges here. M£ are Software producers - theives from Zerox ideas!!! - PSC and the Progress product are a complete different entity. It is improper to compare the products. Ha Ha Thud - laughing my head off at M$ Access. Progress (PSC) are competitors with Oracle, M£ doesn't even enter the arena. Progress rules. hurry up and catch up M$ !!!!! This Millennium!!!

 

[joey.jeremiah]

For argument sake, If you went to Oracle and asked them, well what are you going to do about the Progress competition ?! they'll probably scratch their head get a funny confused look in their eyes and ask who's Progress ? then they'll probably go around the offices and ask who's heard of it and endup googling it, "hey guys I found it !"

I don't think Progress even shows up on the radar in this case


It's hard to say it, but who here believes the 4GL will still be here 10 years from now ?

Although if Progress wanted to move to Java/SQL, maybe they're in a good place to do so with the latest SQL advancements, DataDirect data conectivity and the new Eclipse IDE

But then we'll probably also get rid of the database at some point

 

[joey.jeremiah]

One thing I would like to add though is that -

I think DWP and latest Dynamics release in 10.1A frameworks supports .NET UI rendering besides the regular web UI rendering. I've also heard that a few companies have already built their own using the UI rendering manager

You can find a webevent @psdn.com that shows how the same program you built in Windows can also be rendered to a web UI without the need to do any changes. With the exception that if you have custom client-side coding you will need to write equivalent code that can run in the web browser like JavaScript

I don't know if Java or .NET can do that ?


Something has to be done about the native 4GL User Interface ...
sounds like a serious discussion maybe we should bring it up @PEG

 

[joey.jeremiah]

Dave,

I don't have time for a long post, but my understanding of Progress current direction is there is a recognition that it can't hope to compete with Microsoft and associated 3rd Party vendors on the front end - er- front.

By strengthening its back-end services, and enhancing its integration facilities, it can stay alive in a market otherwise dominated by a couple of big players.

At Progress core is its 4GL, which marks it out from the rest. Using this to provide business services, and looking to (say) Microsoft for the front end seems the way to go for people like me. I will miss ADM2 though...

Also remember that Eclipse will be taking over from Appbuilder in the long run.

Lee

Hi Lee,

I think, the idea behind OERA and BE's and DAO's is that you can use it for a Progress ( WebClient ), .NET, Java, WebServices etc. interfaces, either just one of them or all of them together and with a mix of databases and unmanaged datastore e.g. XML, flat files, webservices etc. seamlessly at the other end

You can also keep ADM/Dynamics as the interface in the same way they've been used.

Though by the time I learn it they'll probably comeup with something else. I'm giving serious thought to becoming a chicken farmer

 

[joey.jeremiah]

I'm in the office ... so here's a code snippet from the _adeevnt.p that we use. We work with Dynamics so you might need to make some small changes. I wrote this junk myself

<snippet>

DEFINE VAR cSourceDir AS CHAR NO-UNDO.
DEFINE VAR cCompileDir AS CHAR NO-UNDO.
DEFINE VAR cBackupDir AS CHAR NO-UNDO.
DEFINE VAR iBackupCnt AS INT NO-UNDO.
DEFINE VAR cSourceFile AS CHAR NO-UNDO.
DEFINE VAR cTargetFile AS CHAR NO-UNDO.
DEFINE VAR i AS INT NO-UNDO.

IF LOGICAL( DYNAMIC-FUNCTION( "isICFRunning" ) ) THEN DO:

IF p_event = "save" THEN DO:

ASSIGN
cSourceDir = DYNAMIC-FUNCTION( "getSessionParam", "AB_source_code_directory" )
cBackupDir = DYNAMIC-FUNCTION( "getSessionParam", "AB_backup_code_directory" )
iBackupCnt = INT( DYNAMIC-FUNCTION( "getSessionParam", "AB_backup_version_cnt" ) )
cSourceFile = SEARCH( p_other ).

IF cSourceFile BEGINS cSourceDir THEN DO:

cTargetFile = cSourceFile + ".v".
SUBSTR( cTargetFile, 1, LENGTH( cSourceDir ) ) = cBackupDir.

RUN verifyDir( SUBSTR( cTargetFile, 1, R-INDEX( cTargetFile, "\" ) ) ).

IF SEARCH( cTargetFile + STRING( iBackupCnt ) ) NE ? THEN
OS-DELETE VALUE( cTargetFile + STRING( iBackupCnt ) ).

DO i = iBackupCnt - 1 TO 1 BY -1:

IF SEARCH( cTargetFile + STRING( i ) ) NE ? THEN
OS-RENAME
VALUE( cTargetFile + STRING( i ) )
VALUE( cTargetFile + STRING( i + 1 ) ).

END. /* do i */

OS-COPY VALUE( cSourceFile ) VALUE( cTargetFile + "1" ).

END. /* cSourceFile begins cSourceDir */

END. /* save */

ELSE
IF p_event = "compile" THEN DO:

ASSIGN
cSourceDir = DYNAMIC-FUNCTION( "getSessionParam", "AB_source_code_directory" )
cCompileDir = DYNAMIC-FUNCTION( "getSessionParam", "AB_compile_into_directory" ).

cSourceFile = SEARCH( p_other ).
SUBSTR( cSourceFile, R-INDEX( cSourceFile, "." ) ) = ".r".

IF cSourceFile BEGINS cSourceDir THEN DO:

cTargetFile = cSourceFile.
SUBSTR( cTargetFile, 1, LENGTH( cSourceDir ) ) = cCompileDir.

RUN verifyDir( SUBSTR( cTargetFile, 1, R-INDEX( cTargetFile, "\" ) ) ).
OS-COPY VALUE( cSourceFile ) VALUE( cTargetFile ).
OS-DELETE VALUE( cSourceFile ).

END. /* cSourceFile begins cSourceDir */

END. /* compile */

END. /* isICFRunning */



/* Creates directory if it does not exist */

PROCEDURE verifyDir:

DEFINE INPUT PARAM pcDir AS CHAR NO-UNDO.
DEFINE VAR cCurrDir AS CHAR NO-UNDO.
DEFINE VAR lExistDir AS LOG NO-UNDO.
DEFINE VAR i AS INT NO-UNDO.

/* File-info was used because search( ) does not work with dir names */

FILE-INFO:FILE-NAME = pcDir.
IF FILE-INFO:FULL-PATHNAME EQ ? THEN DO:

ASSIGN
cCurrDir = ENTRY( 1, pcDir, "\" )
lExistDir = YES.

DO i = 2 TO NUM-ENTRIES( pcDir, "\" ):

cCurrDir = cCurrDir + "\" + ENTRY( i, pcDir, "\" ).

IF lExistDir THEN DO:
FILE-INFO:FILE-NAME = cCurrDir.
lExistDir = ( FILE-INFO:FULL-PATHNAME NE ? ).
END. /* lExist */

IF NOT lExistDir THEN
OS-CREATE-DIR VALUE( cCurrDir ).

END. /* i = 2 to num-entries */

END. /* full-pathnames = ? */

END PROCEDURE. /* verifyDir */

</snippet>

 

[Progress_Guy]

Phew thanks guys.

Joey, agree about backups, I do a midday one and a nightly one, but, unfortunately this was before my midday one. The change to allow saving versioning saves is brilliant. Must admit, never heard of it before.

Adding to my original whinge though it is a crying shame that the ADM is probably going to all 'but disappear' (my opinion only folks - don't shoot me down) as the ability to integrate directly with the 4GL and not have to faff around using ADO and various other bolt-ons is not an attractive one.

The ADM is generally good, but, it is the little things like corrruption, lack of integration and general 'tweaness' that seriously threaten its existence.

Oh well ... where's my .net manual

I'll have a look at the presentation from Progress though on Eclipse.

 

[joey.jeremiah]

applications are not isolated islands, it's never going to be just oracle or sap or java or what ever else. with the oera the business logic is open for other language and applications .net, java, as a webservice, and ofourse still progress. essentially it's the basic 3-tier distributed architecture with a couple more abstraction layers

progress doesn't have to be a server-side language you can still use the webclient, adm/dynamics and dwp as the interface in the same way they've been used alongside with integration to other interfaces and applications

the oera is actually simpler to learn then adm/dynamics. it also takes out some of the bloatiness of adm/dynamics by making some of it's features a core functionality instead of a 4gl implementation

take a look at the docs, presentations and webevents @psdn.com
just to stay up to date with what's happening, HTH

 

[Lee Curzon]

"You can add a small script to the adecomm/_adeevnt.p ( ADE integration hook ) to save back N number of versions whenever you save anywhere in the ADE editor/appbuilder"

I like that. However, I prefer the Syncback option - it's easier to implement, and there are no problems when you come to upgrade your progress version.


"Progress (PSC) are competitors with Oracle, M£ doesn't even enter the arena. Progress rules. hurry up and catch up M$ !!!!!"

I'm sure Microsoft are quaking in their boot disks. Remember MS is responsible for the premier front-end development environment, and a very respected back-end database (SQL Server). We like Progress here, but you are not giving MS enough credit for a very comprehensive development architecture.


"It's hard to say it, but who here believes the 4GL will still be here 10 years from now ?"

I see no other reason for staying with Progress. The 4GL is what sets it apart. Five years ago I figured Progress was on its way out, but I was wrong then, and it seems to have found a renewed charge with V10.


"You can also keep ADM/Dynamics as the interface in the same way they've been used."

You mean while using .Net as a frontend? I certainly hope so, but I also like Client side SmartObjects, and I don't see how they can be integrated; I mean some sort of dual .Net/ADM2 Application front end. .Net for the pretty bits, ADM2 for the quick-to-assemble components.

Of course, if Progress provides some way to quickly link a .Net grid with an SDO, with the grid/SDO automatically linked and updated 2 ways like a SmartBrower, I'm all for that.


"progress doesn't have to be a server-side language"

Surely it does - how can it compete with .Net in the long run?


Lee

 

[joey.jeremiah]

"You can add a small script to the adecomm/_adeevnt.p ( ADE integration hook ) to save back N number of versions whenever you save anywhere in the ADE editor/appbuilder"

I like that. However, I prefer the Syncback option - it's easier to implement, and there are no problems when you come to upgrade your progress version.
Lee

the same as you would when upgrading any piece of code.
compile it and in this case also place it before adecomm in the propath

the ade integration hook as the name implies
is there for exactly those cases

but ofcourse there are other great solutions around
what works best for you ...

"It's hard to say it, but who here believes the 4GL will still be here 10 years from now ?"

I see no other reason for staying with Progress. The 4GL is what sets it apart. Five years ago I figured Progress was on its way out, but I was wrong then, and it seems to have found a renewed charge with V10.
Lee

i was also having a "bad hairday"

"You can also keep ADM/Dynamics as the interface in the same way they've been used."

You mean while using .Net as a frontend? I certainly hope so, but I also like Client side SmartObjects, and I don't see how they can be integrated; I mean some sort of dual .Net/ADM2 Application front end. .Net for the pretty bits, ADM2 for the quick-to-assemble components.

Of course, if Progress provides some way to quickly link a .Net grid with an SDO, with the grid/SDO automatically linked and updated 2 ways like a SmartBrower, I'm all for that.
Lee

i was just making a point that you can still use native progress and
adm/dynamics as the interface in much the same way

one big difference is that instead of using sdo's/sbo's
you'll be using prodatasets which are simpler and more powerful

"progress doesn't have to be a server-side language"

Surely it does - how can it compete with .Net in the long run?
Lee

oera is about integration not .net
interfaces are all about the short run and what's trendy now

e.g. here in israel all goverment offices have moved to linux and openoffice
which also has far reaching implications on affecting the standard of other companies

maybe 5 years from now linux will be the standard in corporate environments ?

 

[vdennis]

Hang in there. The word on the streets is there is going to be a new interface down the pike. I've seen the very early version of it and it will be cool. Ever hear of Eclipse? That is were PSC is heading to.
-Dennis-

 

[Casper]

Hi,

Eclips (in OE10.1) is not about interface but about the IDE.
Since eclips has lots of plugins its nice to have an IDE which integrates seamless with other technology's.
Progress is looking at making OS native interfaces using SWT, but it's still not clear if and when this is going to happen.

Casper.

 

[joey.jeremiah]

H Capser,

> Progress is looking at making OS native interfaces using SWT, but it's still
> not clear if and when this is going to happen.

Really ? I've heard the idea thrown around by someone on PEG, but no one from PSC. TIA

 

[Casper]

Hi Joey,

I heard that in a PUG meeting on UI.
Dave Olson (PSC) told that they're aware of the short comings of the Ui as it is now, and that they're looking in solutions like usng SWT for creating OS native interfaces. If this is going to happen then it's not until R11.

It all depends on the priority the users and PSC have. If all Progress users insist on a better 4GL UI then it probably is going to happen, but that also means that there is less time for other (also) important stuff.

For version 10.1 and beyond the scope seems more DB related then Interface related.
Scheduled are:
referential integrity
On line add/delete field/Index in existing tables
More OO components. (In 10.1 overloading is not possible for instance)

With proxygen and introduction of datasets and new variable types, the 4GL is more open for integration with other interfaces.

Also interesting is the development of windows (longhorn) and xaml. XAML is a way in which you can control precompiled .NET interfaces using XML. (www.xaml.net)

I still think that for most companies who use progress it would be nice to keep building interfaces using the language they know best, so I reaaly hope this SWT stuff is getting somewhere.....

Rgeards,

Casper.