HP3000-L Archives

November 1995, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jim Wowchuk <[log in to unmask]>
Reply To:
Jim Wowchuk <[log in to unmask]>
Date:
Fri, 3 Nov 1995 15:02:01 +1100
Content-Type:
text/plain
Parts/Attachments:
text/plain (98 lines)
At 04:27 PM 2/11/95 -0800, Duane Percox wrote:
>I was sitting here just wondering what plans folks had for utilizing DCE in
>their enterprise (where at least one 3k would exist). Not wanting to just sit
>waiting for periodic DCE posts I decided to ask (what a concept!).
>
>What's cooking with DCE?
>
>1. using it now
 
Nope.  Can't justify the costs for internal use, nor can I see a demand
within the 3000 market for developing externals using it (right now).
 
>2. plan on using it in the future
 
Would like to kick the tyres.  For the moment, RPC is the most appreciated
feature, but there's more opportunity with NT/Server <-> Win95 RPC protocols
(as much as I detest NT Server development) at a *significantly* lower cost.
 
>3. don't know what it is
>4. will never use it (because...)
>
>Impressions on DCE?
>
>1. will it really work for you
 
In danger of being passed over.
 
>2. limitations that bug you
 
Can't find clear statements regarding security algorithm usage outside of
the US.  My fault though...I'm waiting for the news to hit me, rather than
surfing the web and hunting for it.  Its probably common knowledge
(somewhere!).  I certainly would have appreciated having common clocks
between our systems (DCE has a common time module, necessary for kerebos
support)
 
>3. bugs that limit you
 
The runtime for the 3000 should be included with the 3000 (as it apparantly
is with the 9000 UX 10.0).  "Build it and they shall come!" has brought many
disappointments to HP before, but the inverse "Don't build it and they won't
come" is *certainly* true. HP has presumably already done the work to get it
to the 3000 - if it fails to take off, then all that work is lost.  If the
3000 commercial developers can sell products based on it, but as is, the
runtime just becomes another price bugbear not found on other platforms.  On
the other hand, if commercial product providers using DCE find the 3000
supports it, it may make the platform that little bit more attractive to them.
 
Random thoughts:
 
* DCE seems well thought out, though the ONC RPC mechanism was more popular.
Not enough practical experience to see if it works well.
 
* HP has a big involvement in the DCE committee (it was HP's NCS that formed
the basis of the RPC mechanism adopted).
 
* Microsoft has a compatible RPC mechanism, but uses different routines for
identifying services and hosts.
 
* Some experts view DCE as an interim protocol which will pass away once the
true object passing (object-oriented techniques) are common.  HP and IBM
have agreed to the SOM protocol, while Microsoft is going somewhere else.
Gavin recently pointed out a source of interesting information at Xerox Parc
on InterLanguage Unification.
 
* OSF, the organization behind DCE hasn't had great successes before - OSF/1
was supposed to become *the* standard for Unix.  I think DEC and IBM (maybe
even HP) had commercial versions of it out, but I certainly wouldn't call
them the standard.  Did OSF develop the networking standard to replace TCP, OSI?
 
* Microsoft is becoming more and more like the old IBM - if its not made by
us, then buy it - if we can't buy it, then rubbish it.  Sometimes even
Microsoft can't control what they unleash.  Witness the explosion in VBX's
available, many for languages other than Visual Basic (Delphi, dBase, Q&A,
CA-Objects, etc).  Now Microsoft want everyone to move to OCX, but the
market is hanging with VBXs.  NetDDE didn't fly but their next
network/object/procedure protocol probably will.
 
* The range of Groupware (Notes, TeamWare, etc) apps are presenting their
own client/server protocols that effectively accomplish the same things as
DCE is attempting to do.
 
* The 3000 has the stability and integrity to be type of system we all would
wish to see as DCE servers.  Whether its got the performance
(forking/threads) remains to be seen.  But it is certainly a category of
application that I would have expected the 3000 to excel in.
 
>Thanks in advance for your posts (assuming there are any)...
Your wellcome, assuming you think any of this has value :)
 
----
Jim "seMPEr" Wowchuk
Vanguard Computer Services     Internet:    [log in to unmask]
 _--_|\                        Compu$erve:  100036,106
/      \                       Post:        PO Box 18, North Ryde, NSW 2113
\.--.__/ <---Sydney NSW        Phone:       +61 (2) 888-9688
      v      Australia         Fax:         +61 (2) 888-3056

ATOM RSS1 RSS2