* Re: [Caml-list] OcamlSpread 0.0.1 released
@ 2002-09-20 16:14 MONIN Jean-Francois FTRD/DTL/LAN
2002-09-20 16:18 ` Re[2]: " Yurii A. Rashkovskii
2002-09-20 18:57 ` Arnaud SAHUGUET
0 siblings, 2 replies; 5+ messages in thread
From: MONIN Jean-Francois FTRD/DTL/LAN @ 2002-09-20 16:14 UTC (permalink / raw)
To: Arnaud SAHUGUET; +Cc: Xavier Leroy, Yurii A. Rashkovskii, Ohad Rodeh, caml-list
[-- Attachment #1: Type: text/plain, Size: 1178 bytes --]
On a first reading I tended to agree with some
of your arguments. However:
> When choosing a crypto package, there are a few points to consider:
> [...]
> - the maintenance of the package
> Flaws are being discovered everyday. It is better to use a crypto
package
> which is widely used, tested and maintained.
> [...]
> I think the last and worst thing to do is to re-implement some crypto
from
> scratch.
But what do you think of:
"When choosing an operating system, there are a few points to consider:
[...]
- the maintenance of the OS
Flaws are being discovered everyday. It is better to use an OS
which is widely used, tested and maintained.
[...]
I think the last and worst thing to do is to re-implement some OS from
scratch."
More seriously, perhaps, a well-known source of flaws is complexity.
Expressing algorithms in Ocaml is often (much) less complex than
in C...
I cannot be more specific in this case, because I did not examine
the implementations. However, as you say yourself:
> When choosing a crypto package, there are a few points to consider:
>
> - the people who implement the package
Regards,
Jean-Francois
[-- Attachment #2: Type: text/html, Size: 2354 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re[2]: [Caml-list] OcamlSpread 0.0.1 released
2002-09-20 16:14 [Caml-list] OcamlSpread 0.0.1 released MONIN Jean-Francois FTRD/DTL/LAN
@ 2002-09-20 16:18 ` Yurii A. Rashkovskii
2002-09-20 18:57 ` Arnaud SAHUGUET
1 sibling, 0 replies; 5+ messages in thread
From: Yurii A. Rashkovskii @ 2002-09-20 16:18 UTC (permalink / raw)
To: MONIN Jean-Francois FTRD/DTL/LAN
Cc: Arnaud SAHUGUET, Xavier Leroy, Ohad Rodeh, caml-list
Hello MONIN,
Friday, September 20, 2002, 7:14:04 PM, you wrote:
MJFFDL> More seriously, perhaps, a well-known source of flaws is complexity.
MJFFDL> Expressing algorithms in Ocaml is often (much) less complex than
MJFFDL> in C...
AFAIR, alogorithms for cryptokit are implemented in C.
--
Best regards,
Yurii mailto:yrashk@openeas.org
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Caml-list] OcamlSpread 0.0.1 released
2002-09-20 16:14 [Caml-list] OcamlSpread 0.0.1 released MONIN Jean-Francois FTRD/DTL/LAN
2002-09-20 16:18 ` Re[2]: " Yurii A. Rashkovskii
@ 2002-09-20 18:57 ` Arnaud SAHUGUET
1 sibling, 0 replies; 5+ messages in thread
From: Arnaud SAHUGUET @ 2002-09-20 18:57 UTC (permalink / raw)
To: MONIN Jean-Francois FTRD/DTL/LAN
Cc: Xavier Leroy, Yurii A. Rashkovskii, Ohad Rodeh, caml-list
[-- Attachment #1: Type: text/plain, Size: 799 bytes --]
Re: [Caml-list] OcamlSpread 0.0.1 releasedImplementing crypto packages is a real pain.
I wrote an implementation of DES in Java in 1996. Absolutely not interesting for me.
For this kind of stuff, I think writing an oCaml wrapper on top of C library is better.
>
More seriously, perhaps, a well-known source of flaws is complexity.
Expressing algorithms in Ocaml is often (much) less complex than
in C...
-- That's probably true for public key cryptography. For symmetric key crypto and hash function, the algorithms are not elegant.
In companies, there is always the dilemma: make vs buy.
I think the oCaml community should think about "re-implement vs wrap". I don't think the real value of oCaml programming sits at the level of the OS or the low-level libraries.
Arnaud
[-- Attachment #2: Type: text/html, Size: 1739 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: Re[2]: [Caml-list] OcamlSpread 0.0.1 released
@ 2002-09-18 16:13 Ohad Rodeh
0 siblings, 0 replies; 5+ messages in thread
From: Ohad Rodeh @ 2002-09-18 16:13 UTC (permalink / raw)
To: John Gerard Malecki; +Cc: caml-list
I'm the maintainer of Ensemble, so my judgement on which system is better
is of
course subjective. Having said that, I'll try to fairly describe what
Ensemble and
Spread can do.
Generally speaking, they are the same. Both provide group-communication to
users.
Spread splits functionality between a client process, and a server (daemon)
process.
The daemon process needs to run on every machine where there are clients.
User-applications link with the client library, and send/receive messages
through the
local daemon process.
Ensemble allows both client/server, and in-server applications, so that one
can also
write the applications to work as part of the server main-loop. Since
group-communication
applications are normally event-driven, linking into an existing main-loop
should not be
a problem.
Both systems support secure groups, though they have different models for
this. As an
aside, Ensemble intends to move from using OpenSSL to Xavier's cryptokit.
Ensemble splits the "server" into many small layers, where each layer is a
state-machine.
An Ensemble stack is, much like the OS communication stack, a set of layers
stacked on
top of each other. An Ensemble endpoint uses a stack to send/receive
messages. The
set of layers in the stack determine what guaranties are provided:
reliability, flow-control,
packet fragmentation, etc. Users can write they own layers, and there is a
lot of investment
in allowing users to dynamically switch layers on the fly, add new layers,
and debug them.
Equivalent functionality does not exist in Spread.
Ensemble also allows reliable point-to-point messaging, in the context of a
group. This
is similar toTCP, though with somewhat stronger guaranties.
In general, if you want to write an application that does not heavily use
group-communication,
you probably don't care so much about which system to use. If you'd like to
modify the system,
or write something really efficient, I'd suggest using Ensemble.
Ohad.
-----------------------------------------------------------------------------------
Ohad Rodeh
tel: +972-3-6401641
IBM Haifa, storage research
John Gerard
Malecki To: "Yurii A. Rashkovskii" <yrashk@openeas.org>
<johnm@artisan.co cc: Ohad Rodeh <orodeh@alum.cs.huji.ac.il>
m> Subject: RE: Re[2]: [Caml-list] OcamlSpread 0.0.1 released
18/09/2002 06:43
Please respond to
"John Gerard
Malecki"
Yurii A. Rashkovskii wrote (2002-09-18T12:16:01+0300):
>
> I know about Ensemble, and even use it now. OcamlSpread was an old
> library and I just decided to release it to public (may be somebody
> will need it :-) )
Hello Yurii (and Ohad),
Just out of curiosity, and without wanting to start any public
controversy, I was wondering about the differences between Spread and
Ensemble? (Except for implementation language they seem similar.)
Are there applications where one would be preferable to the other?
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [Caml-list] OcamlSpread 0.0.1 released
@ 2002-09-17 20:06 Ohad Rodeh
2002-09-18 9:16 ` Re[2]: " Yurii A. Rashkovskii
0 siblings, 1 reply; 5+ messages in thread
From: Ohad Rodeh @ 2002-09-17 20:06 UTC (permalink / raw)
To: Yurii A. Rashkovskii; +Cc: caml-list
Just to let you know that there is a Group-Communication system written
natively in Caml:
Ensemble. www.cs.cornell.edu/Info/Projects/Ensemble. The latests release
(1.38) is
dated a month ago.
Ohad.
-----------------------------------------------------------------------------------
Ohad Rodeh
tel: +972-3-6401641
IBM Haifa, storage research
"Yurii A. Rashkovskii"
<yrashk@openeas.org> To: caml-list@inria.fr
Sent by: cc: hump@pauillac.inria.fr
owner-caml-list@pauill Subject: [Caml-list] OcamlSpread 0.0.1 released
ac.inria.fr
17/09/2002 04:15
Please respond to
"Yurii A. Rashkovskii"
OcamlSpread, a wrapper for a Spread (http://www.spread.org/) group
communication toolkit has a first release today.
It is quite inmature (it doesn't implement all of the kinds of
functions provided by Spread now) and probably has a couple of bugs.
At this moment OcamlSpread is distributed under the terms of GNU GPL
but will be GNU LGPL later (with notice of Spread license, too)
WARNING: This release should not be used in production.
BTW, I don't spend a lot of time to code it now (last change was about
a month ago) so contributors and/or new maintainer are welcome.
Homepage URL: http://ocamlspread.sourceforge.net/
--
Best regards,
Yurii mailto:yrashk@openeas.org
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives:
http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ:
http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2002-09-23 4:42 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-09-20 16:14 [Caml-list] OcamlSpread 0.0.1 released MONIN Jean-Francois FTRD/DTL/LAN
2002-09-20 16:18 ` Re[2]: " Yurii A. Rashkovskii
2002-09-20 18:57 ` Arnaud SAHUGUET
-- strict thread matches above, loose matches on Subject: below --
2002-09-18 16:13 Re[2]: " Ohad Rodeh
2002-09-17 20:06 Ohad Rodeh
2002-09-18 9:16 ` Re[2]: " Yurii A. Rashkovskii
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox