memory and VM issues

David A. Smith davidasmith at bellsouth.net
Sat Jul 16 02:18:34 UTC 2005


There was some thought given to these ideas in the initial conception of 
Croquet. In a sense, the intent was to be as open as possible, and still 
achieve our goals. For example, there were a number of reasons we chose 
to go with OpenGL instead of D3D or a meta-renderer:

- OpenGL was relatively open and available.
- OpenGL was available cross-platform.
- The controllers of OpenGL were unlikely to "discipline" us, should we 
be successful.
- Since we could provide direct access to the OpenGL API by 
standardizing on it, we are able to easily track and utilize its forward 
progress without major reimplementations of the low-level system.

The same appears to be true of OpenAL and ODE. We are unlikely to have 
to worry about arbitrary third party decisions to control or limit what 
we are intent on doing. This goes to the heart of what we are trying to 
provide here. A system that does not seek to control or limit what you 
can do with it. If we are successful, Croquet will be a platform for 
innovation.

Consider that the Macintosh was first made available to the world in 
February 1984, and Windows followed relatively soon after. This was 
before some of the people on this list were born. Note also that there 
is very little difference between those systems of 21 years ago and what 
we have today. This can be attributed to two factors:

- The interface that was originally developed by Alan and his team 
(leveraging Engelbart's efforts as well) was astonishingly good.
- The part of the interface copied by Apple and Microsoft was placed 
into amber - frozen in its capabilities and intent, and we developers 
were granted only the right to build on top of this interface - not 
extend it orthogonally.

The message here is that this document centric wysiwyg interface is 
simply the best possible interface, and there is no reason for a mere 
developer to consider modifying, extending, or replacing it in any way. 
Consider it to be the analog to the interface to the car. You can't move 
the brake or accelerator, or the next user won't know how to drive it 
anymore. This is true to an extent, but is not the full story. Consider 
that the game consoles have a completely different interface, and allow 
the end user a huge degree of capabilities. This demonstrates that 
alternatives can exist and should be explored.

The reason Croquet is being developed and made available is to allow us 
to begin a true exploration of media. Our understanding and use of media 
is probably the thing that most defines us. We can either be controlled 
by it, as the Microsofts and RIAA of the world would have it, or we can 
be the ones in charge.  It is interesting to note that the subtitle of 
Marshall McLuhan's infuential book "Understanding Media" is "The 
Extensions of Man".



Josh Gargus wrote:

> Uh-oh, OFFTOPIC!!!
>
> On Jul 15, 2005, at 7:18 PM, Alan Grimes wrote:
>
>> Josh Gargus wrote:
>>
>>
>>> On Jul 15, 2005, at 3:10 PM, Alan Grimes wrote:
>>>
>>
>>
>>>> Re:Closed Source drivers.
>>>>
>>
>>
>>>> Just Say No(tm).
>>>>
>>>
>>>
>>
>>
>>> Why, exactly?
>>>
>>
>>
>> Your reasoning makes sense if you consider only the present. It is  true
>> that my sale accounts for 1/100,000,000 sales...
>>
>> But you need to keep your eye on the future.
>>
>
> I do my best.  One thing to remember is that the future is full of  
> exponential curves (or exponentially accelerating
> exponential rates of growth, if some futurist models are to be  
> believed).  Whatever small change might be "achieved"
> by 0.1% of the population not buying this or that brand of immoral/ 
> counter-revolutionary/un-kosher technology will be
> entirely lost in the chaos.  Now is the time for neophiles to ride  
> the changes to the future that they envision, not to dig
> in your heels and try to stop the system.
>
>> In the next decade, computer technology will permiate everything you
>> own. Without the ability to write your own software, all of your
>> posessions will actually be owned and controlled by whomever controlls
>> the code. (NewLine and Microsoft and whomever has power over such  
>> things..)
>>
>> A decade after that things will be worse.
>>
>> If you are a transhumanist, like I am, closed source drivers will be
>> utterly intolerable.
>> There are many schools of transhumanism, not all of which I agree  with,
>> but the basic premice is to enhance the human body and mind with
>> technological augmentation.
>>
>> My philosophy is to draw the line and hold it. At all costs, and at
>> every opportunity.
>>
>
> This philosophy is orthogonal to transhumanism.  For example, it is  
> held by neo-luddite punks.  Not saying
> that it is wrong, just orthogonal.
>
>>
>>> There would be no Croquet project if we followed this  advice.   Such a
>>> policy of abstinence will not change the world; your impact on the
>>> market will not be any  greater than if you simply join the billions
>>> on the planet who don't have access to computers.
>>>
>>
>>
>> =(
>> I am painfully aware of that.
>> I bought 4 AMD CPUs in the last few years to, in part, help support
>> competition in the PC industry... Now AMD's on the ropes and Intel is
>> hitting below the belt and not letting up..
>>
>> AMD bet the farm on Opteron and if Intel hadn't bribed all of the OEMs
>> they would have won...
>>
>
> Maybe, but that's the world that we live in, and it's nothing to be = 
> ( about.
>
>>
>>> IMHO, it's best not to restrict yourself, and instead take advantage
>>> of what is currently available if you want to change the status quo.
>>>
>>
>>
>> Come again? How does that help change the status quo?
>>
>>
>
> Well, for instance, if Croquet gets a billion users (cross your  
> fingers), then the status quo will have been changed (if this  
> happens, widely
> available open hardware will be a more feasible goal to pursue than  
> it is now).  If the Croquet developers decide not to develop Croquet
> because they refuse to use closed-source drivers, then the status quo  
> will remain.  Seems reasonable to me.
>
> Cheers,
> Josh
>
>
>
>




More information about the Squeak-dev mailing list