Hi DexOS !
DexOS wrote:Its hard to put your finger on it, but i get the impression that RISC OS people are stuck in there ways and things much be done in a set way, or not at all.
This has been good in the past and has kept RISC OS alive.
That's probably why it worked and why they did it. Remember RISC OS originally was a commercial product, it was also one burned into ROM so it had to "reliable" as it took effort (and cost money) to fix/change things.
While it might
be argued that as RISC OS has become a more open model that does not remove the need for co-ordination - in fact it probably requires it even more. RISC OS is over 25 years old yet there is only two forks (how many forks of Linux are there ?). Linux can afford it in that it has many enthusiastic and very skilled developers available to it - RISC OS is less fortunate (we have too few developers - so a burst of "forking" would only spread the number of active developers too thin (IMHO)).
DexOS wrote:But now they have a opportunity to take RISC OS to the next stage, but the thing that has kept RISC OS alive is now holding it back.
Possibly, but if it had died out with the last Iyonix back in the early noughties we'd have nothing now. The fact we're having this debate is a good
thing - because it means people can see
that RISC OS has a future - but we just haven't worked out the roadmap fully yet.
DexOS wrote:There's is a need out there for a fast OS that lets coders and experimenters make projects without getting in the way.
It must be small enough to understand, but still have the needed functionality.
Its must not be a linux clone, by slowly moving over to C, there's nothing wrong with linux, i use it everyday, but RISC OS needs to be different.
I fully agree. I would tolerate a little slippage towards "C" because currently there are too few developers and those that ARE doing the development are (I think) entitled to have a say on how they do it. My big issue, however, is that if substantially more than 50% of RISC OS were to become C we'd lose out on the speed and compactness of ARM code. If the % of C got high enough you'd wind up with something slow, that lacks RISC OS'es traditional nippiness and doesn't provide any of the advantages of Linux (you'd in fact wind up - I feel - with a "worst of all worlds" scenario).
DexOS wrote:I think RISC OS need to stay asm and basic, and new programmer's need to be encouraged more.
I'd agree, but would tentatively suggest that some applications (not the OS itself - but tools used on the desktop) could be created using "C" without necessarily compromising things too much. Many apps already are - but key OS modules (filing systems, font system, kernel etc.,) should as much as possible stay ARM coded.
But as I don't have the last say on this (I ain't coding the stuff) I'd make that and my other suggestions as tentatively as possible as the ROOL lads have done a remarkable amount of work - and it shows in that RISC OS has become a viable, fast and interesting OS on the RaspberryPi and other platforms.
DexOS wrote:I for example have spent a long time coding a usb stack in full assembly for my hobby OS for the pi, now why am i doing that when RISC OS is my ideal OS ?.
Some progress has been made on USB on RISC OS (even support for USB Audio devices...) so perhaps things are better there than they were (and yes the USB situation on RO was not the best and is improving slowly). Perhaps what you've learned could be of use in further advancing USB on RISC OS ?
DexOS wrote:Its the little things that need to be looked at, for example where on any of the main forum's is there's a programmers sub forum ?.
A good point.
DexOS wrote:Please note, this is just a outsider looking in point of view and the people that have worked on RISC OS have done a great job.
Again I am in full agreement with you they have indeed done a great job
Thanks for your considered feedback !