Unified ROM Project, Phase II: The Kitchen Application - Page 2

**schettj** · #11 (**permalink**) 04-04-2007, 06:56 PM

I think if you want to get commercial apps in your rom, you should be building it yourself...

Re: the other question of building the rom off the kitchen based on user web interaction. Yeah. Not trivial.

The only rom tools I know of are windows based, although they used to be linux based tools maybe. If they were linux based, building a kitchen as part of a cgi script (ie, based on filling out a form and clicking the BUILD button) wouldn't be too bad, except for the huge load it will put on the machine and the slight hairy issue of dealing with multiple requests at the same time.

If we're stuck with the windows tools we have now, then
* someone has to run a windows-based web server or provide network APIs to drive the tools,
** some of which need to have buttons clicked to proceed,
** none of which have great error reporting if the user selects too many OEMs and the ROM is garbage because it has too much in it,
* you can forget processing multiple build requests at the same time (so, we would need to batch them up, build in series, and maybe email the user a URL to fetch the ROM once its done?)

Icky.

pgdeaner · #12 (**permalink**) 04-04-2007, 07:32 PM

I liked the original kitchen concept.

Whatever the final product is it should be small, self contained, and simple to use.

What would make the kitchen useful would be a wealth of OEMs that could be added or subtracted, perhaps even commercial apps with the .exe files missing.

I think most people like myself with some technical experience were able to build ROMs using the existing kitchen. The problem was creating OEMs that worked and did not conflict with other OEMs. That would be a useful tool. maybe a ".cab to OEM" converter? That would be sweet.

RussMuscle · #13 (**permalink**) 04-05-2007, 01:49 AM

I would prefer the kitchen myself. I run apps that no else on this forum seems to run, (at least none of the ones I use have been mentioned). I created the OEM's for them for the 3.3 kitchen and would like to integrate them into the 3.5 ROM, which is running great on my phone. It would just be easier for me to have the kitchen.

Just my .02

**willysp** · #14 (**permalink**) 04-05-2007, 09:34 AM

Guys,

I'm a bit confused. I've been assuming all along that at some point I'll be able to download a kitchen to my computer and work with it just like I've been working with the earlier kitchens.

I hope that what you're talking about is an additional way for people to use the kitchen.

We'll still be able to download a 'stand-alone' kitchen, correct?

Thanks again for all the work.

**jamesus** · #15 (**permalink**) 04-05-2007, 11:09 AM

Quote:

Originally Posted by willysp

Guys,

I'm a bit confused. I've been assuming all along that at some point I'll be able to download a kitchen to my computer and work with it just like I've been working with the earlier kitchens.

I hope that what you're talking about is an additional way for people to use the kitchen.

We'll still be able to download a 'stand-alone' kitchen, correct?

Thanks again for all the work.

Yes, there will still be a stand alone kitchen that any user can cook their own ROM. This is simply an additional tool for the novice user.

Quote:

Originally Posted by pgdeaner

What would make the kitchen useful would be a wealth of OEMs that could be added or subtracted, perhaps even commercial apps with the .exe files missing.

Great idea!

Quote:

Originally Posted by pgdeaner

I think most people like myself with some technical experience were able to build ROMs using the existing kitchen. The problem was creating OEMs that worked and did not conflict with other OEMs. That would be a useful tool. maybe a ".cab to OEM" converter? That would be sweet.

Not sure of its feasibility, but a .cab to OEM convert would be awesome!

**colonel** · #16 (**permalink**) 04-05-2007, 12:02 PM

A cab to OEM converter would be quite nifty, but would take a feat of programming. Unfortunately, cabs are not always that simple. The trick lies in where the cab creator puts the registry entries. Most cabs just have the registry entries as part of the cab or a simple file such as a .reg. These types of cabs are easy to OEM. The problem comes about when there are self-registering .dlls or Setup.dlls that add registry entries upon install. So, someone would need to create either a process tracker to track the created reg entries or some sort of disassembler to be able to view the self-registering dlls.