At 11:54 6/2/01 -0800, James Duncan Davidson wrote:
>On 2/2/01 12:56 PM, "Peter Donald" <[EMAIL PROTECTED]> wrote:
>
>> Consider the case of select() style functionality - why has not that been
>> added. It's not because it isn't portable, nor is it because it is
technically
>> hard or a design challenge. I put it to you that the sole reason is
because it
>> would mean that low end servers with a small number of CPUs could actually
>> start to compete with the suns network OS hardware which would be a bad
thing
>> for them. Yet not having this functionality has proved considerably
>> challenging to anyone running java as server and in native threads mode. I
>> suspect this will change with Merlin release as more people were allowed to
>> have an opinion.
>
>Oh this is just hilarious. Of course you aren't going to believe me since my
>motives are always so "questionable", but the reason it's not there is
>because the JDK team have been working through their queue of real problems
>in the platform and this hasn't yet burbled to the top of their queue. It's
>not that it's not a real problem, it is, but when you have finite resources
>and need to structure them to get things done, triage results in things
>dropping that are unfortunate.
>
>Now, whether or not the model that is being used is efficient for getting
>work done or an appropriate model is not something that I want to get into a
>discussion on -- but I'm simply pointing out that you are ascribing to
>malice something that isn't the result of malice.
It makes good buisness sense to not implement select() and is not the
result of malice...
Let me see - which step did I stuff up.
1. Sun coined or at least advocates the term "The network is the computer"
2. For a while Java has been billed as great for serverside stuff
3. The longest bug/feature request relating to serverside stuff is lack of
select()
4. Implementing select() is not difficult considering it is part of POSIX
and implemented on all platforms that I am aware of (except small ones
which use a different platform - J2ME).
5. select() was implemented by other vendors hence the R&D costs should be
minimal
6. Without select it is impossible to write scalable server apps that deal
with sparsly transmitted upon connections except in hardware that have fine
grain locking + many CPUs
7. One of suns greatest revenue streams is selling hardware that has fine
grain locking and many CPUs
8. Given the above - Sun obviously believes networking is important, server
products are a forte of java and have little cost or risk implementing
select().
9. By virtue of 6 and 7 it would be an unwise buisness decision to
implement select() as it would reduce demand for their hardware.
Conclusion ?? (it should be obvious).
Now which step in reasoning is invalid?
Cheers,
Pete
*------------------------------------------------------*
| "Computers are useless. They can only give you |
| answers." - Pablo Picasso |
*------------------------------------------------------*
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
