[mjc]

I'm in the design stage of creating an unattend app to demolish all other unattend apps.

It will contain each of the unique features of nLite, RyanVM's updates, BTS's driver packs, XPize, PowerPacker, MiniPE, and any other tools people can recommend.

I'm being paid to do this as part of my 9-5 because we need such a tool internally, but I will make the program freely available, under the GPL if licensing allows.

It is evolving out of a script that I've been working on for a few weeks. I've decided that 100% of the process beyond selecting what is needed and not needed can be automated and therefore should.

Current features include:

• Accepts any Windows XP or 2003 version set (will do more than one version at a time soon)
  
• Downloads and slipstreams the appropriate service pack
  
• Downloads and installs the latest RyanVM pack (this should instead be downloading and installing the latest hotfixes directly from MS in the future)
  
• Applies nLite tweaks (nLite is currently used for slipstreaming and ryanvm anyways, but I'd like to roll all of these functionalities into one giant app)
  
• Applies BTS driver packs (should download all the drivers separately in parallel from each manufacturer)
  
• Creates RIS installs (with license key management to come.. different .sif files for each key, permissions set to the user, computer, or group policy that is allotted the keys)
  
• Creates a WinPE disk (to come; with preferred tools only as an option)

The app will be written in C, because it is the language I am most comfortable in, and because it is the fastest for such things.

I plan on including the functionality of burnatonce and cdimage.exe as well. cdimage.exe's important function (optimization) is very easy to implement.

If you are interested in working on this project with me, let me know.

I would appreciate keeping feature requests and whatnot to a separate thread as I expect this one to grow fast.

[Nanaki]

It should rely on other people's software (why recreate something?), but with the permission of those other people, of course. If it combines them with an easy-to-use GUI, it would rule.

*awaits first screen or release*

keep it up!

edit: implement cdswitch

This post has been edited by Nanaki: 05 July 2005 - 01:48 PM

[mjc]

Nanaki, on Jul 5 2005, 03:47 PM, said:

It should rely on other people's software (why recreate something?), but with the permission of those other people, of course. If it combines them with an easy-to-use GUI, it would rule.

*awaits first screen or release*

keep it up!

edit: implement cdswitch
<{POST_SNAPBACK}>

the problem with using other peoples' software is speed. Hand-crafted and hand-optimized C can be tons faster than simply launching a bunch of separate tools.
If I am able to release it (in whole or part) under the GPL, not only would it be faster, but the source would be modifiable by all. (think: no need to create another tool when you can just extend AIOG)

[Nanaki]

Don't agree. People like Nuhi or Bashrat have been working quite some time on most of these programs, so why recreate their work (probably less good due to less time). Also, the programs have been written with several languages (nLite runs on .NET for example). I understand your goal is everyone starts writing in C. :/

[mjc]

Nanaki, on Jul 5 2005, 03:56 PM, said:

Don't agree. People like Nuhi or Bashrat have been working quite some time on most of these programs, so why recreate their work (probably less good due to less time). Also, the programs have been written with several languages (nLite runs on .NET for example). I understand your goal is everyone starts writing in C. :/
<{POST_SNAPBACK}>

if nLite is in C#, porting it to C is trivial.

Most of the code in most of these applications is actually very straightforward, and with some thinking and access to the source, it would be very easy to port and/or integrate all the features.

I'm trying to reduce the amount of work done in the future by unifying all these tools. If it takes a lot of porting, rethinking, retooling, or whatnot, so what?

Also, as I'm being paid to create this, I have all the time in the world to crank out updates, fixes, manuals, and whatnot.

The only licensing restrictions that would prevent me from publishing the whole or part of the tool would be restrictions by the creators of the current toolset, as well as restrictions on Microsoft or other vendors' code (for example the driversets...)

[mjc]

mjc, on Jul 5 2005, 04:03 PM, said:

Also, as I'm being paid to create this, I have all the time in the world to crank out updates, fixes, manuals, and whatnot.

To clarify, we need a RIS tool, I'm creating one. No commercial fishiness involved. I own the copyright to my own code, so I'm free to release it under the GPL.

[nuhi]

hehe to demolish all other.

Good luck, you'll need it.

[dirtwarrior]

This would rule. All I have to say is good luck,and I would like to test your product

[glent]

This is just my opinion and dont mean to be negative but:

you will not impress many members of this forum with your plans you wish to rethink your plans and maybe considering help the devolpers out in a different way or altleast discusing it with them

This post has been edited by glent: 05 July 2005 - 02:46 PM

[mjc]

glent, on Jul 5 2005, 04:43 PM, said:

This is just my opinion and dont mean to be negative but:

you will not impress many members of this forum with your plans you wish to rethink your plans and maybe considering help the devolpers out in a different way or altleast discusing it with them 
<{POST_SNAPBACK}>

I plan on implementing this tool, as there is a need at the company I work for, regardless of whether I have aid from others or not.

As far as discussion, that's what this thread is for =)

Developers can choose to support me or do their own thing, whatever works. I'm not forcing anyone to do anything.

As far as rethinking, if you mean because it would obsolete a lot of work, I feel that this is the best path to producing the product we've all been waiting for =) if you feel otherwise please explain how and why.

[Nanaki]

Good luck then... you'll need it.

[Mike_Wilson]

you make your own bed, u can lie in it

[XPero]

Good luck but I dont bet for your project. Also, I'm not with stealing other people ideas but...Looking forward the first public build.

[RyanVM]

Have fun

[mjc]

XPero, on Jul 5 2005, 04:53 PM, said:

Good luck but I dont bet for your project. Also, I'm not with stealing other people ideas but...Looking forward the first public build.
<{POST_SNAPBACK}>

I would by no means consider writing such an app stealing. After all, credit will be given where credit is due...

Without the help of everyone that's ever put work into this stuff I certainly would never have dreamed up such an app.

[smknight]

Very ambitious. It will be awesome if it lives up to the intention.

People shouldn't be hostile for the potential this will make other programs obsolete. Down the line results will speak for themselves and people will use the program or set of programs that they feel the best is for them.

[mjc]

mjc, on Jul 5 2005, 10:13 PM, said:

seeing as the work is already largely done, if I had enough cooperation from {nuhi, ryan, etc) it would really not be too hard.  my timelimit is 6 months anyways.. so tis not a weekend hack..
<{POST_SNAPBACK}>

Cooperation meaning help of any kind (programming, docs, source, candy...) of course. By no means do I expect anyone to just sit back and watch if they do not want to.

[Alanoll]

I wouldn't hold your breath on getting help from many of them. They'll answer questions i'm sure, but i doubt they'll give you detailed information on how to recreate what they did (or in the case of nuhi, opening up the source).

[Mike_Wilson]

There is no reason they should rly, they have spent alot of time creating programs such as nlite etc. You cant expect them to just give it over.

[bevone]

will you paid for doing that?? that's great,
I'm a C/C++ programmer also, but actually I can test my nLite installation only in 23-01,