Memory hogging in an OS without paging

Question about which tools to use, bugs, the best way to implement a function, etc should go here. Don't forget to see if your question is answered in the wiki first! When in doubt post here.
User avatar
Candy
Member
Member
Posts: 3882
Joined: Tue Oct 17, 2006 11:33 pm
Location: Eindhoven

Re:Memory hogging in an OS without paging

Post by Candy »

kmcguire wrote: I figured instead of defining a global standard for communication, I decided on a way that would be suited for the service or driver, I think you could call it. It just felt like too much work. It sounds like you are making a global encapsulation standard which is great, but is it worth it and what are the advantages you get from it?
If I can find a way to make a very modular non-application specific method of processing data, that I can make applications work better and more efficient without reloading the entire application, your entire system would be able to get a certain thingy working if you install it once and so on. Discoupling applications with a certain interface and object type it can work on with the actual code that does the change, formatting, effect, input, output or anything.

My computer: OpenOffice, Acrobat Reader, notepad, wordpad and so on. Why? Because acrobat can read PDF's (which the rest can't), OO can make complex documents (which the rest can't), Notepad has low overhead (which the rest isn't) and Wordpad can handle unix enters with medium overhead (dito).

I only have acrobat reader because it's required for viewing PDF files. What about changing that into a module that converts PDF into class Document? There goes Acrobat reader.

What about not loading modules when I don't use them? There comes the speed. Now, I can just pick any interface I like and it'll have the same exact functionality as all the others, so I'm choosing an interface because I like the interface.

If I don't like the interfaces available, I can make a new one without reproducing the effort to read PDF files or to write ODT files. I can just assume it's there and be done with it. I can spread my interface since I'm not statically linking with any other module (so the modules can be GPL and my interface PD, or the other way around, or proprietary modules such as Adobe's PDF reader etc.).

I have Winamp, Windows Media Player, Mplayer, Media Player Classic, RealPlayer (although I've kicked it off again) and quicktime, all for some reason. They all do the same, but in a slightly different way. Screw that, just the interface I want with the input/output from any of the formats.

I believe that I have made the technology to do so. I am only requiring time to make my OS work that far and to create enough drivers so it'll be useful.

Stretch the idea, make a web application from which you can get modules for anything you want. You get a list for "solutions" of one or more modules that give you a certain conversion you want, including full price and so on. You choose what module is used for what, what you don't install and so on. The licensing is per user, with optional online section hosting for your private data, although you can use USB sticks for that as well. You have your settings in that section.

End result: You can use any computer, use the software you've paid for (whether it's on your computer or any other), use it with your preferences and so on. You set everything right once and it stays that way. You don't have any maintenance (or very low, updating some modules) and can just use your computer for what it was intended.
I am it seems, if I understand you correctly, going to need to (from the eyes of a developers) write modules that dynamicly link
Yep.
, or the equivilent which is on my mind for starters a library of obj and headers that the program includes that are the CRT you could say for communication with a certain service.
(I know CRT isnt the right word, but I was trying to stay consistant with my last post. :D)
(The process<->kernel comm. is contained in the CRT.)
Each has the same interface up to a certain level. Every normal application (and some of the more awkward ones) are based on two things:

- Data
- Transforming data

Data is in specified types. Transforming data is done with modules that you can request through a library and that takes an input data object (or multiple ones, or a stream, or none) and that outputs an output data object (or more, or a stream, or none).

This abstraction works very good for the applications I need on my OS. You can make a word processor, mp3 player, file exchange software, IM clients and all other stuff you would want. For some programs it can't replace the entire program, but it can make it a lot easier to write. There's just one slight thing: It cannot work together with DRM, security-through-obscurity or such. There's no obscurity and you can't tell the output program what to not do with data.

If it becomes popular, I'm in for a fight with the RIAA, MPAA, Brein (dutch thingy) and so on.
User avatar
Kevin McGuire
Member
Member
Posts: 843
Joined: Tue Nov 09, 2004 12:00 am
Location: United States
Contact:

Re:Memory hogging in an OS without paging

Post by Kevin McGuire »

I see. I understand! :D
JoeKayzA

Re:Memory hogging in an OS without paging

Post by JoeKayzA »

@Candy: Wow, that looks very interesting! I thought about a design with similar goals some time ago, but I never went that deep...

The idea was to have modules (just like you do) which provide some functionality, to form an application, you put them together and describe this in an XML-file (I was inspired by J2EE in this regard). Furthermore, UI-forms should be designed in XML too (in a similar fashion to xhtml), so the UI and logic parts of an application are always seperated, and new UI's can be designed relatively easily (and then just plugged into an application). I think you get the idea. :D

cheers Joe
Post Reply