Hello,

quoted from Michael Gelman at Megido-general ml:

>>>>>>>>>>>>>>

Since I am the only currently active Israely team member, i can tell the following:
1) :-( the israeli team may be going apart.
2) :-) the server will be still supported FOREVER.
3) :-) we still have ALL the stuff necessary to continue.
4) :-) Does the "lack of authority" really need to bother us? Anyone who
wants to become a leader | do sthg useful is INVITED to do so - CVS, may I
remind you, is working, as well as everything else... The ideas are worked
out, compiler team is alive and ready, gtk is still out thee, and ~100 members can contribute. So what's your problem - we are BAZAAR, for tux's
sake!

>>>>>>>>>>>>>>>

A little late to realize! :-(

greetings

Nico

Search for your X11 directory and add it to your path. It seems that it's not in your search path...

Shane

I agree we need the editor, but for now I guess I would like to see others get involved creating controls and not worry too much about Lazarus compile problem. I'm sure the problem is simply the way that the procedures are defined. I'll look into to as soon as possible.

I also believe that we DO need a list of tasks. I'll give you a start. On Cliff's page (www.pcpros.net/~vbman )we have a short list of who is working on what. Along with that on my page (www.lakefield.net/~smiller/lazarus) we have a list of controls. This is pretty accurate but should be refined. There aren't a ton of controls written yet so it would be easy enough to generate a list.

Cliff, does your simple IDE work now?

John, I will look at the comile prob and get back to you.

Shane

As much as I HATE to do this...

Me Too.....

Is it possible someone could drop an executable in my Inbox ? Pretty please...

I'm using the old 0.99.11 compiler off the lazarus web page, but there are problems compiling gtk at the moment, with this compiler. Tho, gtk will probably compile with the latest CVS compiler, lazarus won't.

Thanks

Brad....

"Out of my mind, Back in 5 Minutes"
"Windows Multitasking: messing up several things at once" "Microsoft. How do you want to crash today?"

/"\
\ /
X ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
/ \

In a message dated 6/24/99 4:09:08 PM Eastern Daylight Time, writes:

<< The problem why lazarus doesn't compile anymore is that procedures are assigned to methodpointers which are not compatible. But methodpointers are in the compiler only for two months now and are still under (heavy) development. >>
Even when I comment out all the offending methodpointers , I get another error saying something like:
ld -lX11 file does not exist
I'm at work now, I'll post the actual error from home if necessary. I would like to help out, and if I could get a taste of Lazarus it may help me see what I want to work on.
-Angles

Do not get me wrong here I am not dropping from for Megido I think they already folded.

Cliff

I agree we need the editor, but for now I guess I would like to see others get involved creating controls and not worry too much about Lazarus compile problem. I'm sure the problem is simply the way that the procedures are defined. I'll look into to as soon as possible.

The problem why lazarus doesn't compile anymore is that procedures are assigned to methodpointers which are not compatible. But methodpointers are in the compiler only for two months now and are still under (heavy) development.

Peter

Hi all. I've been lurking for a while, but decided it was time to speak.

Borland are releasing J++ Builder for Linux second quarter of next year, after a Solaris release late this year or early next. I saw a screen shot of it running on Linux+GNOME at a recent Borland roadshow here in Adelaide. They also talked a bit about maybe porting C++ Builder, but there were no definite plans.

Having said that, I wouldn't count on Java GUI's being fast enough for usage in most projects for another few years, esp. in large projects.

I've done some work in it, and while Java is great for server side applets, in most client applications the GUI speed is slower than I remember Windows 3.1 being on my old 386.

Before GNOME & KDE came out (say back in '95-97), there seemed to be a general feeling in the Linux community that there was no need to work on a good GUI, because Java would get there soon. Since then, GNOME and KDE have both been written from scratch, and either make a better choice for a desktop environment at the moment that Java (IMHO).

The other thing is (and I may be wrong here) that 'Java in the Kernel' isn't really happening. What the Kernel allows you to do (and has done for a while) is allows you to run any binary format you have a interpreter. You still need your JVM as a separate program. Correct me if I am wrong about this.

Anyway, I'm looking forward to the day when I can use Lazarus to program under Linux the just have to recompile them for Win32.

Hello Everyone...

I have been deeply in some work and besides the weather has been absolutely great. (Makes it harder to sit behind a PC)

Myself and Shane live in Wisconsin that means that we only see about three months of summer before the weather gets ugly.

I sort of blame myself for the stall of the project currently but I hope to resume my work in a few weeks.

Personally I wish to keep continuning the effort but in my heart I really believe that Linux GNOME and KDE is going to go away in the future. I have thought much about this and can see a evolution starting to mount.

I believe that real soon someone is going to figure out that it is not a big deal to embed a JVM Java Virtual Machine into the linux kernel. Well this can already be done with the latest relase but soon we will see someone start distributing it. Now that is a really good idea and wish that I had the resources and time to do it myself. Think about it no GUI stuff to build it already is there. Java is so much like the Pascal Languge it is almost theft. Not to mention all of the net support that is built into it.

Just sharing a few of my thoughts..

Not gone just taking a little break guys. Myself and Shane worked 24/7 for about 4 months on what has already been built. Once things calm down for me a little I will jump back in.

The newest compiler has been released "Cool"

Cliff

I believe you're correct, Ron. And that's even beside the point, as far as I'm concerned. I've never liked java because the apps seem sluggish and running the jvm makes my box sluggish. People that know much more than me won't touch java for those and other reasons. I for one would never compile the jvm into my kernal because I don't need the weight.

I'm trying to develop a cross platform gui app using fpc/gtk. It would definately be easier to use java but I'm biting the bullet now for a compact, high performance app down the road. Ever since I made the jump from vb to delphi several years ago I've felt this way and I'm not turning back.

Cliff: I hope you give this some more thought. Lazarus is a great idea and making some headway, the past month notwithstanding. An open souce Delphi-like tool that will build cross-platform apps is the way of the future, in my opinion. You might also want to test the waters for general opinion of Java - many people aren't fans.

Keep up the good work!

Mike

I have two questions:

1. Anybody know when asserts will be supported by the compiler? I feel naked without them. (I have the 0.99.12 release, but I still get "Out of memory" errors when I turn on {$ASSERTIONS+} ).

2. I still can't compile the Lazarus source! I still get the method/procedure errors in dialogs.pp. I got the source direct from CVS today in hopes that it would work. If this is still not ready, could somebody email me an executable? Send an i386 or PII linux executable to

Thanks!

John Margaglione

_______________________________________________________________ Get Free Email and Do More On The Web. Visit http://www.msn.com

There is a IDE Open Source called FreeBuilder that runs on linux.

Cliff

I just wanted to add, I have heard of efforts to bring java development environments over to Linux as a separate product .. stuff like Borland's J++Builder and stuff like that. I dont recall if Borland was actually going to do it, (I think they rumoured it). Anyway, undoubtedly a GUI Java Dev environment would be ideal for many people, especially an open source one. I agree with Mike on the speed issue -- as far as a tight object-pascal compiler, it just seems like un-neccessary overhead.

Unfortunately, due to my job's demands I do not have the time to participate in the development process of this project -- however I do like to throw in my 2 cents every once in a while :)

== Ron

There is only a few of use doing any work on the Lazarus code. I think what has happened is that for a variety of reasons we have all hit a snag due to other work that has kept us from doing much the last several weeks. I myself hope to untangle a little time starting next week to get back into it. More so now that v12 is officially released.

As for myself when I get into it I will be working more on the messaging between GTK and FPC. I am trying to separate as much of the FCL from the API that is being used so that it will be easier to use the FCL with say with Gnome, QT, xWindows, etc.

I'm all for the list of what to do and who's doing it.

So you can be put me down as working on the FCL-GTK interface.

It should work, please sent some code which doesn't work

The lazarus sources are not updated since Jun 1, so the bugs are still there.

Peter

Things are a bit stalled. I have blown my Linux system up and have been using my other system to test Windows 2000. I will get back to it shortly. The problem in lazarus isn't a major thing. See, Lazarus is a side issue, it's the components and controls that need to be built. Lazarus is going to be an IDE once those components and controls exist. I think it would be a good idea for those interested to find a GTK widget that we currently don't have and create it, or clean up our current controls. No one seems to be creating controls at the moment other than myself and I am stalled.

Simply go to gtk.org, look at the widgets they have, look into out code and see how we are doing it (look at tbutton and trace it to tcontrol to see how it's done) and add a new control. Notify the list when you plan on doing a gtk widget so know one else does it.....please I beg you! I need help creating the controls.....

Shane

Brad Campbell el d�a Wed, 23 Jun 1999 11:17:50 +0800, escribi�:

I've sended an updated dialogs.pp weeks ago, but I guess it get lost in the mailing list.

Anyway, things seems stalled...
(and I'm doing others things)

Sergio

John,

I program Java and Delphi and I agree with what you've said. Although I've done alot more Delphi than Java, your absolutely right that it's a *NIGHTMARE* to do a good Java GUI. In particular, I find the use of Layout Managers to be a VERY time consuming endevour. I use JBuilder and its more or less forcing me to purchase a new machine for home. My Pentium 233MMX with 128MB RAM is just too slow. You really can't do JBuilder Java GUI developement without at least a high-end Pentium II machine. You need a minimum of 128MB RAM, I recommend 256MB, especially if your interested in developing Java Servlets.

I'm new to Linux, Free Pascal, GTK and Lazarus. I am moving this week, but as soon as I settle in I'll drive into the details and see what it's all about. Hopefully, I can find an area in which I can contribute.

Do you think the Lazarus project can coax the Megido Project folks to completely bail out and join Lazarus? You (Lazarus) guys seem MUCH more together than the Megido Team. It would seem to make sense to focus efforts. Pascal Forever!

Java Problem #4: Sun is NOT doing a good job of promoting third party tools and reusable components for Java. They are not very good at encouraging independent Java development partners. Everything new in JavaLand is centered around a new SUN Java API. Like you said, the APIs are powerful, but Java will never be as easy to develop in (especially GUIs) as Visual Basic or Delphi until Sun stops monopolizing all the innovation. We can all bad-mouth Microsoft and Bill Gates, but MS is excellent at developer and third party relations.

I'm interested in researching how to develop cross platform apps using the Free Pascal Compiler. Is it possible now? Or only after the Lazarus project is further along?

That's all for now...

Tom Wilk

Message text written by

As someone who has two shelves full of Java books and more Java compilers

and IDE's than Bill Joy, I can honestly say that Java will not be going anywhere soon. Try as they might, they will never get compiled speed for Java in the *GUI* arena. Not unless they make swing a bunch of DLL's that internally call X or GTK. And I find that highly unlikely. Java also has a
number of other problems that will keep it at bay for several years:

1. The language is simple, but the construction of programs (especially when considering swing) is not. It takes hundreds of lines of Java to get a
JTable working well. Besides the sheer lines of code, the way in which AWT

works, with adapters and listeners and so forth, plus Swing's pluggable architecture, is a *nightmare* for GUI IDE designers. Take any Java IDE out
there and you will see the same thing: poor integration with Swing. VisualAge, my favorite tool, is still locked into Swing 1.0.3. JBuilder can't redraw anything the same way twice. Visual Cafe has very little "visual" to it.

2. The best part of Java right now is the APIs. As brutal as the amount of
code can be, I love the API's. They are powerful and flexible. But NONE of
it can't be done in a well written Pascal API. You said yourself that Java is a rip-off of Pascal. Well, you're right. And if you really want a challenge, instead of rewriting the VCL, rewrite the JCL (Java Class Libraries, not Job Control Language :).

3. Microsoft hates Java about as much as it hates Linux. Enough said.

Anyway, Java is still a long way from being ubiquitous. We have ample opportunity to shape Lazarus into whatever we want. If the VCL is outdated,
then don't implement it. Want adapters on top of the VCL? Go for it. FPC

compiles as fast as Java, is as clean a language (but boy what I wouldn't give for the compiler guys to implement checked exceptions), and GTK isn't going anywhere soon. A good abstraction of the VCL/whatever will make us plenty independent of GTK anyway.

Besides, we do this for the joy of it, not for profit. If something better

comes along in a few months/years, we can jump on board that ship. In the meantime we can make a really nice IDE.

Besides, I just spent the better part of the day figuring out the Pascal grammar in hopes of writing the Object Browser/Intellisense portion of Lazarus. I'm going to be bummed if nobody is around to use it!

John Margaglione

"Buck up little camper, don't let Stalin get to ya!" -from "Better Off Dead"
<

As someone who has two shelves full of Java books and more Java compilers and IDE's than Bill Joy, I can honestly say that Java will not be going anywhere soon. Try as they might, they will never get compiled speed for Java in the *GUI* arena. Not unless they make swing a bunch of DLL's that internally call X or GTK. And I find that highly unlikely. Java also has a number of other problems that will keep it at bay for several years:

1. The language is simple, but the construction of programs (especially when considering swing) is not. It takes hundreds of lines of Java to get a JTable working well. Besides the sheer lines of code, the way in which AWT works, with adapters and listeners and so forth, plus Swing's pluggable architecture, is a *nightmare* for GUI IDE designers. Take any Java IDE out there and you will see the same thing: poor integration with Swing. VisualAge, my favorite tool, is still locked into Swing 1.0.3. JBuilder can't redraw anything the same way twice. Visual Cafe has very little "visual" to it.

2. The best part of Java right now is the APIs. As brutal as the amount of code can be, I love the API's. They are powerful and flexible. But NONE of it can't be done in a well written Pascal API. You said yourself that Java is a rip-off of Pascal. Well, you're right. And if you really want a challenge, instead of rewriting the VCL, rewrite the JCL (Java Class Libraries, not Job Control Language :).

3. Microsoft hates Java about as much as it hates Linux. Enough said.

Anyway, Java is still a long way from being ubiquitous. We have ample opportunity to shape Lazarus into whatever we want. If the VCL is outdated, then don't implement it. Want adapters on top of the VCL? Go for it. FPC compiles as fast as Java, is as clean a language (but boy what I wouldn't give for the compiler guys to implement checked exceptions), and GTK isn't going anywhere soon. A good abstraction of the VCL/whatever will make us plenty independent of GTK anyway.

Besides, we do this for the joy of it, not for profit. If something better comes along in a few months/years, we can jump on board that ship. In the meantime we can make a really nice IDE.

Besides, I just spent the better part of the day figuring out the Pascal grammar in hopes of writing the Object Browser/Intellisense portion of Lazarus. I'm going to be bummed if nobody is around to use it!

John Margaglione

"Buck up little camper, don't let Stalin get to ya!" -from "Better Off Dead"

_______________________________________________________________ Get Free Email and Do More On The Web. Visit http://www.msn.com

I believe that some people have mastered the technique of "pre-loading" the JVM and keeping it in memory, making it faster to initially load a program from the command line.

Doesn't make Java GUI faster enough though.... and second the "good for serviettes and applets, but bad for GUI" idea. (Having worked for a large Java oriented Software Company for 14 months now.

you wrote

I'm interested in researching how to develop cross platform apps using the Free Pascal Compiler. Is it possible now? Or only after the Lazarus project is further along?

Tom:

FPC has compilers for many platforms. GTK has been ported to win32 (there's a working version of gimp for win32) via a set of dll's. The fpc gui examples which use gtk compile under both win32 and linux, assuming you have the libraries/dll's installed. I got an earlier version of lazarus to compile and run under win32 a few weeks ago. As long as lazarus is careful about the units used (like the linux unit) it can be a cross-platform tool.

In a nutshell, it's possible to use fpc/gtk right now for win32/linux development. When lazarus is more mature, it should be a lot easier.

I've been getting the hang of gtk programming the past week and should have some example code available shortly if anyone wants it.

Mike

I must be missing something. I dont see how a Java Virtual Machine replaces a Desktop Environment such as KDE or GNOME. Now, I'm no Java programmer so i'll take your word that Java does all that GUI Interface stuff coolly.. If I am understanding you correctly, you would want to use java's already implemented GUI interfaces, which are unto themselves and do not rely on KDE or GNOME. My question then is, how do you achieve integration with controls which are configured by KDE, Gnome, or some other desktop environment -- in other words, shouldn't the controls appear to be subclassed from the standard controls provided, with the user's color selection and graphic style?

Maybe I am missing the point, or my lack of technical understanding behind desktop environments like KDE and GNOME -- and how they interact with Java.

== Ron

I agree that the components are important at this stage in the game, but there are other factors to consider.

1. Programmers create out of necessity and frustration. The biggest reason I want Lazarus running is that all the other
editors lack decent object pascal support. I would be much more motivated working on a specific widget if I desperately needed that widget for a dialog in Lazarus.

2. The best way to test the components and Lazarus is to use them side by side. An annoying bug in TButton? It will probably rear its ugly head a lot faster if it keeps you from doing something while you're programming with Lazarus.

3. Team motivation. Everybody likes to have a flashy demo to show to their friends, wife, etc.. Pride drives most programmers, and nothing boosts the old ego like a flashy gui app.

4. I just want it to compile!!!

Anyway, is there a list of what components are being wrapped, which are being worked on by who, etc? It's been so long since anything was checked into CVS that it's kind of hard to tell what's going on anymore. Maybe we could have a project progress page with a list of todo's and responsible parties. I'll draw up the HTML if someone can give me a decent list of tasks.

John Margaglione