62 replies to this topic

[Brito]

Didn't the automatic download of the Windows ISO is now failing? (Microsoft taking it offline)

[Trumk]

On demand I'm ready to add every senceful extention to WinBuilder 083.

 

 

I vote for keeping 083 and adding sensible extensions. Thanks!

[pscEx]

Didn't the automatic download of the Windows ISO is now failing? (Microsoft taking it offline)

No idea. Maybe there is another downlod server. If not, I'm sorry, the user must have the original Win install CD ISO. But that is also neccessary for all existing PE Bulders starting with Bart.

 

Peter

[Trumk]

I know you know this nice place to provide you with direct download links. No spamming intended, credits to Mkuba50

https://mdl-tb.ct8.pl/

Edited by Trumk, 22 February 2017 - 04:09 PM.

[homes32]

Install WinBuilder into a directory of your choice and run it. Your installation is the project. On Build a bootable minimum PE ISO will be crated.

You can extend the PE by your own addons.

 

Peter

and that's my point. i don't want to extend the default PE. I want to make my own. winbuilder was originally built on the freedom of choice to build whatever PE I want

 

if I plug a Win10 source or a WinXP source into the current winbuilder is it going to build me a Win10 or WinXP PE?

 

lock-in to one persons idea of what a PE should be built from and what it should include is EXACTLY why people aren't using it.

 

EDIT:

I'm all for the idea of a default minimal build for those just wanting  to get a look and feel of what can be done (remember nativeEX_barebone?) but it should not restrict others from creating more advanced projects with different sources and it certainly should not be given "special" treatment by the builder with "project only featuers"

Edited by homes32, 22 February 2017 - 04:27 PM.

[Brito]

All of that is possible. The current version you find is pre-configured for a single project. The idea is configuration-zero, just run, press a button and get an ISO.

 

Of course that if you want to make changes, it can also do that.

 

Peter, perhaps we can update the download link to elsewhere and make this one-click work again?

 

I can likely find a permanent location. Microsoft doesn't seem to care any longer about their ISO images.

[homes32]

All of that is possible. The current version you find is pre-configured for a single project. The idea is configuration-zero, just run, press a button and get an ISO.

 

Of course that if you want to make changes, it can also do that.

 

I'm not understanding. (perhaps due to lack of documentation....see a pattern?)

 

the current zero-config win7pe appears to be compiled into the winbuilder.jar so how is it that I can modify it?

how can I add a win10pe project? recompile the whole darn builder? (hint: impossible as nobody but you and psc have the source)

 

remember this?

 

- Personal comment

People have the rigth to know how to make boot disks with their own resources,
and they surely deserve good tools and quality scripts to do them the way
they really need. Each included project can easily be ported to other programs
or tools in order to achieve similar or even better results, there's no
intention to produce closed source projects, our goal is quite the
opposite - everyone is free to understand how everything works and give
a positive contribution to further improve any project available.
 
Let me know if you have a good idea or project you wish to develop using
WinBuilder, there's also a forum dedicated to all sort of ideas where
you can post your thoughts on boot methods. (http://boot-land.net/forums)

As long as you're willing to work hard to get good results, expect nothing
less than my dedicated contribution to help things go forward..


Have fun!!

--nuno brito
 

[Brito]

The intention at the time was to make a minimal Win7PE with as little as possible, and then add/change features or programs on the top.

 

Basically make sure that every time there is a solid base underneath. Has anyone asked before for the source code of that project or wanting to create another one? Because I'd be sure to make it available. Unfortunately there was no such interest. For addons, they were made available on http://reboot.pro/fi...ory/10-plugins/and can be installed directly from winbuilder.

 

This meant that anyone could upload addons and they'd be available. I get the impression we are being criticized for the tool not being perfect. I know very well it is not perfect. It was coded for simplicity and did that task well.

 

Can be improved? Other projects? I'm looking forward to that. In fact, we've learned a lot and could as well do the changes exactly as you mention. At the time it was first released, the challenges were different. And like I've already said, we can and should make the programming/development easier. I can volunteer to make the overall UI similar to the old winbuilder in navigation, now I know better how to get it done.

[homes32]

Basically make sure that every time there is a solid base underneath. Has anyone asked before for the source code of that project or wanting to create another one? Because I'd be sure to make it available. Unfortunately there was no such interest.

I'm not sure if anyone has ever asked for the source. I know people have asked how to make their own project and kept getting blown off and pointed to how to make app scripts.
also because development of wb20xx was more or less behind closed doors so nobody knew how it worked and I'd wager nobody expected projects to be hard coded into the jar/exe. I
I've yet to get a straight answer from either of you how to add a new project.

 

For addons, they were made available on http://reboot.pro/fi...ory/10-plugins/and can be installed directly from winbuilder.
This meant that anyone could upload addons and they'd be available.

this was not always the case and due to lack of communication and no answers to questions from project developers by the time you could drop a script in to run locally devs and lost interest or decided that spending time and effort developing scripts for a "closed" project was too risky and moved on.
 

I get the impression we are being criticized for the tool not being perfect. I know very well it is not perfect. It was coded for simplicity and did that task well.

That's not my intention. I've yet to find any software that is perfect. It may be simple But it certainly isn't developer friendly. and without developers there will be no users.
 

Can be improved? Other projects? I'm looking forward to that. In fact, we've learned a lot and could as well do the changes exactly as you mention. At the time it was first released, the challenges were different. And like I've already said, we can and should make the programming/development easier. I can volunteer to make the overall UI similar to the old winbuilder in navigation, now I know better how to get it done.

l get it. really. its easy to get caught up in a goal and not want to take time to do boring stuff like write docs and tell people how to do things, specifically when things are likely to change rapidly like they can do in early development. but alienating the project dev base by not answering questions, publishing a road-map, bug-tracker, etc did not help the cause. I would love to see winbuilder live on be it in delphi, java, C#, c++, or something else but the community has become so fragmented and there is a definite precieved disconnect between wb devs and project/script devs.  I'm not sure where to start rebuilding. I just what to go back to having fun.

[Trumk]

could as well do the changes exactly as you mention...

we can and should make the programming/development easier.

I can volunteer to make the overall UI similar to the old winbuilder in navigation, now I know better how to get it done.

 

That sounds good!!!

[misty]

...I would love to see winbuilder live on be it in delphi, java, C#, c++, or something else but the community has become so fragmented and there is a definite precieved disconnect between wb devs and project/script devs.  I'm not sure where to start rebuilding. I just what to go back to having fun.

The King is dead, long live the King!

And a particularly big thumbsup to the having fun bit!

Misty

[Wonko the Sane]

The King is dead, long live the King!

And a particularly big thumbsup to the having fun bit!

Misty

Maybe even Rule #12 could be re-instated, as well as all the good ol' ones, and getting rid of the current site policies (inspired by Amnesty International ):
http://reboot.pro/to...82-board-rules/

12. SMILE! Life is tough, we all know that, when you enter this board, it will be appreciated that you leave your personal problems behind, asking and replyinq questions or however exchanging informations with a "positive" attitude, TAKE IT EASY and enjoy your stay here.

 

Wonko

[Brito]

Maybe even Rule #12 could be re-instated

 

Sorry. This kingdom is frozen. Thou shall not have fun around here. Get back to work.

 

 

 

I've yet to get a straight answer from either of you how to add a new project.

 

Realistically, the best way at this moment is open sourcing. It would be possible to extract the .java files and change the extension to .bsh but I have the impression that it is better that everyone can look at the whole engine to see how everything is working (not perfect, but still better than navigating at the dark without full code access).

 

I would put the engine on github or similar and then developers can use the NetBeans IDE to create a new project. Very likely using the default project (win7PE) as an example or first template. The previous guiding principle was to build a solid core project for each platform (win7, win8, win10, ..) and then let this core be customized as desired by users. In the future we can certainly change/improve this as feedback arrives.

 

NetBeans makes scripts programming much easier. Especially to see if something is wrong on the syntax and debugging (step-by-step seeing the value of each variable). I would be adding unit tests. These are automatic tests that verify the output of functions with different inputs, to make sure that a change at some part of the engine doesn't break some functionality elsewhere. The initial ones will be done by me, would ask for help of other members to keep adding tests so that we can verify the output of these functions as rock solid. (and avoid the person A broke on purpose the functionality C type of arguments).

 

Scripts don't need Netbeans. They are standalone and any text editor will do. Albeit ideal is to also use netbeans because then the function auto-complete and syntax check becomes available when writing code. I guess we could make available a developers SDK. Basically a barebones environment for anyone to create a plugin. I can do this as well, should actually have been made available from the start (sorry).

 

How would this look?

[homes32]

Realistically, the best way at this moment is open sourcing. It would be possible to extract the .java files and change the extension to .bsh but I have the impression that it is better that everyone can look at the whole engine to see how everything is working (not perfect, but still better than navigating at the dark without full code access).
 
I would put the engine on github or similar and then developers can use the NetBeans IDE to create a new project. Very likely using the default project (win7PE) as an example or first template. The previous guiding principle was to build a solid core project for each platform (win7, win8, win10, ..) and then let this core be customized as desired by users. In the future we can certainly change/improve this as feedback arrives.
 
NetBeans makes scripts programming much easier. Especially to see if something is wrong on the syntax and debugging (step-by-step seeing the value of each variable). I would be adding unit tests. These are automatic tests that verify the output of functions with different inputs, to make sure that a change at some part of the engine doesn't break some functionality elsewhere. The initial ones will be done by me, would ask for help of other members to keep adding tests so that we can verify the output of these functions as rock solid. (and avoid the person A broke on purpose the functionality C type of arguments).
 
Scripts don't need Netbeans. They are standalone and any text editor will do. Albeit ideal is to also use netbeans because then the function auto-complete and syntax check becomes available when writing code. I guess we could make available a developers SDK. Basically a barebones environment for anyone to create a plugin. I can do this as well, should actually have been made available from the start (sorry).
 
How would this look?

That could work if we can get a few more eyes on things. and the unit tests would defiantly help avoid the issues we had with the 8x branch.
the sdk and GUI are all good ideas as well and would help ease adaptation or a least more people writing scripts to test.

have you tried compiling the current version into machine code? I'm definatly interested in seeing how that plays out rather then haveing that massive runtime sitting around.

[Brito]

have you tried compiling the current version into machine code?

 

Good question. I've done it for other projects and documented the steps here: http://nunobrito1981...native-x86.html

 

We'd later need to test on this specific project (size, speed, startup times) to get some benchmarks. The .JAR version is the largest when we include the JVM (about 60Mb). The .NET compiled version is likely the smallest (at about 10Mb with the IKVM libraries added) and the x86 I'm not sure, wasn't looking at those details when testing.

 

Creating the binaries is relatively simple to automate. We can actually include the compilation tools so that no third-party apps are needed to install and keep this step simple.

[TheHive]

Can be improved? Other projects? I'm looking forward to that. In fact, we've learned a lot and could as well do the changes exactly as you mention. At the time it was first released, the challenges were different. And like I've already said, we can and should make the programming/development easier. I can volunteer to make the overall UI similar to the old winbuilder in navigation, now I know better how to get it done.

 

[misty]

Further development of winbuilder is exciting news (yes I know, I really do need to get out more!).

I do not have any programming skills to contribute, however I'm happy to volunteer some time for testing and contributing to writing any documentation.

Regards,

Misty

[Brito]

Thanks Misty, it is surely a good help (and I'm sure TheHive would also be around for feedback)

 

We need two (or more developers) to help on the engine-side. Homes32, Paraglider? This would likely be the best way to make sure we don't repeat mistakes from the past.

 

Java programming skills are not essential, on the first steps we basically need to make everything easy for everyone (non-java-programmers included) to download the SDK and press the play button to compile a working binary. If it isn't simple for you to get it working, it is already failing and should be improved. As time moves, I think that making minor changes on the code becomes easier until the point you fully understand how to use the language to add new functions. This would also be a good opportunity to understand the engine internals and with that knowledge add your contribute on the kind of engine you would like to see available.

 

On my side I would like to focus the next weeks creating the UI portion while others can work on the engine. But I will also be around when help is needed.

 

Would suggest that we first start by preparing the engine developer SDK so that anyone can join in the future. Then preparing the project developer SDK (adaptation from the previous SDK) and finally go for the script developer SDK. This exercise would already make everyone involved thinking about a common API from the start (for the specific tasks such as creating an icon/link or others). The goal would be that anyone can run the development side of winbuilder without trouble. A bit more of work right now, but at least making things simpler for anyone to join in the future.

 

Once we have the building blocks, should be easier to introduce new projects for newer (or older) versions of Windows.

 

[homes32]

I'll defiantly help as I can. I'm not a java programmer but given enough time to learn how the code works I can learn.

[paraglider]

At the moment work is keeping me very busy. Don't have the time to invest in new projects. Before the new winbuilder can have broader appeal I think the following has to happen:

 

1) It needs a gui

2) You need to change the name and separate it in these forums from the old winbuilder. It has led to everyone being confused.

3) The pe project needs to be separated from the engine in the jar file and be separate downloads(s). Anyone needs to be able to add / edit local app scripts without having to rebuild jar files. Don't know enough about java / bash to know if this is even possible.

4) Engine needs to support downloading app / projects from servers other than this forum server.

5) Commands don't seem obvious. Most people would type help. But that is not a supported command. It should be added as an alias to list. I think the say command should be removed - it provides no useful feature. Also remove the useless quotes before the command prompt.

[Brito]

Let's get working on these items.

 

We still need someone to be in front of this project. Volunteers?

 

Preferably a person with a good interest in seeing the Windows 10 project available. Both myself and Peter wouldn't likely qualify, would be better to have someone who can help organize what needs to be done.

 

[alacran]

@Nuno

 

With all due respect  a subforum for Win10PE should be opened long time ago.

 

I made builds using MistyPE project from Windows 10 on June/23th/2015 (no sysWoW64 for Win10 jet) see: http://reboot.pro/to...e-7#entry193291

 

MistyPE project is capable to make PE builds through the good old Great Winbuilder v082 without download additional requirements (ADK/WAIK), from the following 32 and 64 bits installation media:

• Windows Vista SP1
• Windows 7
• Windows 7 SP1
• Windows 8
• Windows 8.1
• Windows 8.1 (Update 1)
• Windows 10

In 64 bits versions sysWoW64 can optionally be added in the build in order to run 32 bits programs on the PE.

 

The build process has been tested on the following operating systems (from the author old guide, new one in process):

• Windows Vista SP1 (32 and 64-bit)
• Windows 7 SP1 (32-bit and 64-bit)
• Windows 8 (64-bit)
• Windows 8.1 (32 and 64-bit)

 I can personally confirm buidings from XP (SP3) OS too, see: http://reboot.pro/to...e-2#entry178401

 

In last full download sysWoW64 for Windows 10x64  was added: http://reboot.pro/to...e-8#entry202696

 

EDIT: Forgot to say building time From log: Project time: 3 minutes, 15 seconds and 922 milliseconds having all free programs downloaded from a previous build. 

 

Best Regards

 

alacran

[Trumk]

Alacran, I agree MistyPE is great, but surely you wouldn't want to stop Nuno from developing a new builder, would you? Right before it could (almost) see the light of day?

Or are you really just talking about the Win10PE subforum part? If so: yes, there could be a 10 subforum, but at least you did a fine job of promoting MistyPE("-10"), also, Misty maybe doing some more "promotion" for it himself, once theWow64 part is past "beta" and all the project files are fully updated, tested and documented.

Still, nice of you to promote MistyPE!

[alacran]

 

Alacran, I agree MistyPE is great, but surely you wouldn't want to stop Nuno from developing a new builder, would you? Right before it could (almost) see the light of day?

 

NO, I never said that in my post. Who am I to stop new developments?, I'm not againts progress.

 

 

Or are you really just talking about the Win10PE subforum part?

 

YES, It was the primary intention, This Forum has a WinPE-10 capable proyect, and we need to let people know that.

 

 

but at least you did a fine job of promoting MistyPE("-10")

 

I was trying to give support to my statements, anyway, excluding Misty of course, I think nobody else has make more builds than myself, so I know very well what the project is capable to do and how fast it is. So as I do not know if Nuno, PscEX and all the others that are going to work (or maybe they just started) in this new development have tested MistyPE, as I haven't seen posts of them in misty project (there is one from Nuno about problems in the download), then I was trying to let them know better about it's capavilities, I'm quite sure they will find good ideas for their project if they take a look.

I come to the forum usually every day and since several months ago it has been very quiet, I would like to see more good people here, not spammers (this is my school, the little I know, I lerned here and in MSFN), the forum needs more developers (many of them have gone) so a little promotion is not bad and that way we can attract some smart people with good knowledge in programing and at the end it is going to be good for all of us. Anyway I didn't say anything that is not true in my last post the links are there to support what I said.

 

Trumk: I'm glad you make your post with some questions, it allowed me to clarify doubts (if any).

 

Best Regards

 

alacran

[misty]

Thanks all for the publicity and nice feedback. A WinPE10 subforum will be very lonely if MistyPE is the only project in it. Bring on the new builder.

I have lots of suggestions to make for any new project and am more than happy to share the knowledge gained from coding my own projects.

A new project utilising the talent on this forum will add more options for people and may also feed into the development of MistyPE / Mini-WinFE.

More later.



Misty