Hello All,
First post in this thread just read through it - pretty disappointed.
I hasten to add at this point I am LINUX user, have been since 1994, stopped dual booting last century.
Later one of the benefits I can add its to look at the proposed implementations and see if it can be made to work under LINUX, and point out if it cannot before anyone writes a line of code- assuming LINUX is at least desirable as a target audience. I am sure there are people that can do the same on OSX.
Personally, I think the "OS+setup is rubbish at security is derailing the thread", it started off with such hope. I really could not care much less if I tried - and I have, if Windows measures up to some security specification - that is for government and large commercial buyers. We are feeding back on a character generator. Windows users needs to be within the target audience.
The DB argument was more relevant from an implementation point of view, but even that was jumping the gun.
A much better discussion is which Windows OSs and configs to be compatible with - should be go back to ME/98/95?. Still premature in my book.
My thoughts from a CT point of view:
Gal2.4 works well for me for sector mapping. Never have got H & E to work under LINUX.
Online character generators spit out characters for CT.
There is a huge difference to me in pointing my web-browser at a site that is hosted by someone else at their expense and having software to produce an NPC on my computer.
There are actually 2 inputs I can see that have not been addressed, that affect architecture.
1) The level interaction required during character generation.
2) Are we producing a PC generator where the user has control or an NPC generator load up variables - and a character gets generated.
There is a lot of input such as home-world, pre-career etc decisions to make in T5 - but it is focused towards the beginning of the process - there is only one career, so looks a good a candidate for complete automation with variables (i.e. scenario 2 above).
Sorry do not know what someone is trying to produce, so which DB (if any by the way as (2) is stateless), security and the rest of the posts after the first few are jumping to design before requirements.
Of course users make a major difference. However, due to inherent problems with Win registry system & the networking code base there are major exploitable holes that will never go away. For Mac & Linux, you're right. So few people use those systems that it isn't worth peoples time to create malware for them.
Yes and no. There is no real patch available to defend against well constructed "root kits". Due to MS licensing requirements for security vendors with Vista & Win 7 this is more problematic than it was for XP as far as proactive 3rd party protection.
Strictly speaking yes. However there is a design aspect that goes into it. No matter how much testing is done at what cost, Windows will never be at 6 or 7 as there are design criteria at that level. See Orange book for original specs on which CC is based.
