[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: NLDM LDM future... 32 types, server side restriction.

Hi Peter,

Peter Silva wrote:
> >
> You are remarkably quick in responding.  This is quicker than most tech
> support!
> The answers are reassuringly clear as well.
> 

Thanks!  Happy to hear that.


> > My view is that the data relay should work with any NNTP server. As
> >
> >However, the issue of general NNTP compatibility will be visited if/when we 
> >decide to
> >go this route as it's not totally my own decision.
> >
> >
> >
> What is the process to go through with making the decision ?   Is there
> anything pushing people to make a decision, or is it just a matter of
> general planning for improvements to be made as resources permit.
> 

I would characterize it as a little of both.  The pushing piece is the cost of 
doing parallel development of both NLDM and LDM.  Although you may have only 
heard about it recently, it's actually been going on for a while.  We've 
identified a deadline of
the end of March to make a decision based on the information available.  It is 
my goal to test enough aspects of this approach that the decision can be as 
well informed as possible.

The process for making the decision isn't really defined at the moment.  One 
piece of my plan is to consider invididual aspects of each approach and try to 
do a side by side comparison of each.  But, INN has many more features than 
does LDM.  What I also
want to do is experiment with all the new possibilities that INN provides, 
like: automated routing, backlog handling, dynamic construction and destruction 
of connections, pulling articles,  multiple storage buffers of varying types, 
etc.

(The tradeoff - INN is also much more complex than LDM and more difficult to 
administer.)

> We're the official source for feed type 20 (GEM)...   We're a bit snowed
> under (no pun intended) at the moment, but
> sometime in the spring/summer, we might be interested in setting up a
> parallel NLDM feed.  That is, we would keep our LDM server for official
> UNIDATA product delivery, but feed feed the NLDM network from the same
> machine as well.  Are there any conflicts that would arise from this?
> Would that be a useful thing to do?
> 

Offhand I don't see any conflicts from doing that other than the load on the 
machine.  But, as (I think) I said, I'm running both LDM and NLDM on my machine 
without any problems.

Would it be useful? - seems like it could be.  Should we actually transition to 
the new protocol, I guess it would depend on when in the process you want to 
switch over.  Hard for me to say more than that at this point.

Are you at all interested in participating in the test network?  I would love 
to have a location as "remote" (geographically speaking, of course) as yours.   
Just a thought, no pressure... 

Btw, have you seen the NLDM statistics page?  Take a look at 
http://my.unidata.ucar.edu/content/projects/nldm/relayStats/plotStats.php, 
although please bear in mind that the stats are not quite bulletproof and 
you'll see some trace statements and possibly
an error or two.  (Also, the Windows/Java versions are not running at the 
moment - I should remove those.)  If you wanted to participate your stats would 
appear on that page.

Anne


> --
> address@hidden  -  Data Acquisition & Distribution, Environment Canada

-- 
***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                 P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://my.unidata.ucar.edu/
****************************************************