headers
Mark S. Miller | 7 Jun 2004 18:56

Please Help: Recovering Consistency After Crash/Revival and Lost Messages

Hi all,

I'm back, and alive again after a terrible outage. I have much good E news 
to report (soon, on e-lang), and much to catch up on, but first:

Due to an unfortunate mistake I made (while recovering from a Netsky virus 
infection), I have permanently lost all email sent to me between May 23 and 
June 4. If you've sent email to me personally during this time, please 
resend if possible. Thanks.

--

-- 
Text by me above is hereby placed in the public domain

        Cheers,
        --MarkM
Permalink | Reply | 0 comments |
headers
Mark S. Miller | 7 Jun 2004 19:13

Re: edoc

At 04:15 AM 5/17/2004  Monday, James Graves wrote:
>I finally found some time to look through edoc.
>
>Unfortunately, it seems there are only about 6 files which aren't
>tainted by the SCSL.  That doesn't really leave much code left.

If I remember right, the interesting logic of edoc is in those files. 
edoc doesn't have much logic since it doesn't really do that much. Javadoc 
does most of the work.

>It seems it may be better to just re-do the whole thing from scratch. 
>There's a couple main choices for implementation language: Java or E.
>
>If we go the Java route, SinjDoc seems to have progressed a bit
>since MarkM might have last looked at it:
>
>    http://cscott.net/Projects/GJ/#sinjdoc
>
>So we might want to start from that as a base.  It is GPL, BTW.  Is that
>OK, or did we want a less restrictive license?

The 0.5 version of sinjdoc (the current one at the moment) is the one I 
looked at. It looks good to me.

GPL is a potentially serious problem, since the heart of the current 
E-on-Java implementation is covered by MPL, and the two are link 
incompatible (or something). However, this should not prevent us from 
building and running your new edoc as a separate program, which is how I ran 
the old edoc, in order to avoid SCSL contamination.
(Continue reading)

Permalink | Reply | 1 comment |
headers
Mark S. Miller | 7 Jun 2004 19:22

Re: Bug (0.8.26h): EScript.optMethods doesn't opt

At 09:56 PM 5/17/2004  Monday, Kevin Reid wrote:

>? e`pragma.enable("plumbing")
>>   def x match y {}
>>   `.subs()[1].rValue().getEScript().optMethods()
># problem: <NullPointerException>

Good catch. Fixed in the soon-to-be-announced 0.8.28d.

--

-- 
Text by me above is hereby placed in the public domain

        Cheers,
        --MarkM
Permalink | Reply | 0 comments |
headers
Mark S. Miller | 7 Jun 2004 19:26

Re: Bug (0.8.26h): deSrcKit does not handle non-identifier verbs

At 05:45 PM 4/29/2004  Thursday, Kevin Reid wrote:

>? <elib:serial.makeSurgeon>.withSrcKit("").serialize(def x { to __optUncall() :any { [1,
"add(Number)", [1]] } })
># value: "(def t_0 := 1).add(Number)(t_0)"
>
>? <elib:serial.deSrcKit>.makeBuilder().buildCall("1", "add(Number)", ["1"])
># value: "1.add(Number)(1)"
>
>Options:
>
>  1. E.call/3
>  2. e.enable.verb-string

Fixed in the soon-to-be-announced 0.8.28d. As of 0.8.28d, 
e.enable.verb-string is now an official part of the language, so I took 
option #2.

--

-- 
Text by me above is hereby placed in the public domain

        Cheers,
        --MarkM
Permalink | Reply | 0 comments |
headers
Mark S. Miller | 7 Jun 2004 23:16

Pre-Announcing 0.8.28d: Donut-Lab in 72 hours. 3-vat CapTP. An E IOU; always Y...

At http://www.erights.org/download/0-8-28/index.html

Until further testing, this is currently just a release candidate. Due to 
various disasters, I currently have no working Linux or Mac OS X systems, 
and so will not be able to build binary releases for these platforms yet. If 
you have any problems building from sources on these or any platforms, 
please let me know. (After this week, I'll be gone for two weeks, and so 
won't be able to repair the situation until the end of June.)

Note that the last official E release is 0.8.26h. The 0.8.27 series is 
skipped, but remains as a milestone of what we were able to accomplish using 
E in only 72 hours, all of which is included in 0.8.28d: 

                      Donut-Lab: A PlanetLab with no center

PlanetLab http://planet-lab.org was created to serve as a testbed for 
experimenting with large scale distributed systems of various kinds. Among 
the systems now running on PlanetLab are various DHTs, content distribution 
networks (like CoDeeN), proposed DNS replacements, and other highly 
decentralized, massively distributed systems. However, PlanetLab itself is 
an administratively centralized system. "PlanetLab Central" remains a 
central point of failure for all the systems built on top of PlanetLab, 
which is unfortunate, as these systems themselves are often impressive 
exemplars of systems designed not to have such centralized vulnerabilities.

When I raised this issue with Timothy ("Mothy") Roscoe, one of the PlanetLab 
architects (and, btw, an old capabilities guy), he said "PlanetLab is a good 
enough testbed to prototype its successor." Indeed. Rick McGeer, an E fan on 
the PlanetLab steering committee, has gotten E working in a PlanetLab sliver 
(a Linux VServer, which is how PlanetLab nodes currently offer compute
(Continue reading)

Permalink | Reply | 1 comment |
headers
James Graves | 8 Jun 2004 00:04
Favicon
Gravatar

Re: edoc

On Mon, Jun 07, 2004 at 10:13:02AM -0700, Mark S. Miller wrote:

> If I remember right, the interesting logic of edoc is in those files. 
> edoc doesn't have much logic since it doesn't really do that much. Javadoc 
> does most of the work.

OK, I'll have to read through it again.

> The 0.5 version of sinjdoc (the current one at the moment) is the one I 
> looked at. It looks good to me.
> 
> GPL is a potentially serious problem, since the heart of the current 
> E-on-Java implementation is covered by MPL, and the two are link 
> incompatible (or something). However, this should not prevent us from 
> building and running your new edoc as a separate program, which is how I ran 
> the old edoc, in order to avoid SCSL contamination.

[...deleted...]

> Because a javadoc-like program is such a big job, perhaps you should at 
> least do the first prototype using Sinjdoc? Once we have something working, 
> we can reevaluate from there. How's that sound?

OK.  We probably don't have a major need to have it totally integrated.

Unfortunately, I'm hip-deep in seven different projects at work, and the
girlfriend is flying into town this week.  So I'm not likely to get
anything done for a couple weeks, at least.

Welcome back!
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark S. Miller | 8 Jun 2004 16:15

The Birth of the Oz-E Project

[cross posted to e-lang and users@...]

The biggest news from my Belgium trip is the birth of the Oz-E project, led 
by Fred Spiessens. (As part of the MILOS project 
http://renoir.info.ucl.ac.be/twiki/bin/view/INGI/MILOSProject .) Before I 
explain the Oz-E itself, I'll explain the name. It turns out we were already 
converging on a naming convention for applying E ideas to other languages, 
but didn't quite know it yet:

Joe-E
    The application of E ideas to the Java language, in order to create a 
    capability secure distributed language which is as compatible with Java 
    as is reasonably practical. ("Java" is a Sun trademark, but "cup of Joe" 
    is slang for coffee.) See the second half of 
    http://www.eros-os.org/pipermail/e-lang/2004-April/009797.html
E-on-Java
    The current E implementation, in the Java language

Squeak-E
    The application of E ideas to the Squeak language, to create a 
    capability secure distributed Squeak-like language. See
    http://swiki.squeakfoundation.org/squeak-e
E-on-Squeak
    Dean's work in progress to port Kernel-E onto the Squeak virtual machine.

(Along these same lines, if the attempt to apply E ideas to make a secure 
Python is revived, it could be called Pyth-E (pronounced "pithy"), or Mont-E.)

The Oz-E project will attempt to apply E ideas to the Oz language to create 
a secure distributed language that maintains the flavour of Oz, and is as
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark S. Miller | 10 Jun 2004 03:30

Bug in 0.8.28d:

Those using E's persistence should avoid 0.8.28d and wait for 0.8.28e, due 
to a bug MarcS just found, whose symptoms look as follows:

># problem: Can't setReadOnly
>"c:/elang/scripts/donutServices/sugarMint/mint.vat"
>
>#
>#   - static FileSugar#setReadOnly(OneArgFunc)
>#   . <file:c:/elang/scripts/donutServices/sugarMint/mint.vat>.setReadOnly(null)
>
>#    <at>  setReadOnly/1: <jar:file:/C:/elang/e.jar!/org/erights/e/extern/persist/makeAtomicFile.emaker#:span::55:16::55:26>

More soon...

--

-- 
Text by me above is hereby placed in the public domain

        Cheers,
        --MarkM
Permalink | Reply | 501 comments |
headers
fsp | 10 Jun 2004 12:00
Picon

The Birth of the Oz-E Project - Announcement

[cross posted to oz-e, e-lang, users@... and announce@....org]

Mark, thanks for the announcement.

The decision to start with a separate Oz-E project, apart but still very much 
connected to Mozart-Oz, was taken during Mark Miller's visit to Belgium, in 
connection to Fred's "confirmation" proof. We hope this "temporary separation 
of concerns" will enable the Oz-E designers and developers to give complete 
priority on security and protection issues from day one, while still 
maintaining the Mozart-Oz look and feel of the language as much as possible.

People interested in following-up, supporting, contributing and realizing this 
joint-effort Oz-E project "To design and implement an Oz-like secure 
multiparadigm language", are invited to do so. Mark Miller already described 
the initial approach and strategy in his mail, included below. 

We've set up a wiki site for the project, to contain the results from 
discussions on design and implementation issues. It currently has a single  
introduction page, containing a proposal for the mission statement.
Have a look at it here: 

http://renoir.info.ucl.ac.be/twiki/bin/view/OZZY/WebHome

The site will be updated soon, to contain a more complete plan. Suggestions 
are very welcome in this area too! 

To join or follow the Oz-E project discussions, please subscribe on the Oz-E 
mailing list at this address:

     http://mailman.info.ucl.ac.be/mailman/listinfo/oz-e
(Continue reading)

Permalink | Reply | 0 comments |
headers
Peter Van Roy | 10 Jun 2004 16:15
Picon

Final CFP: Second International Mozart/Oz Conference (MOZ 2004)

 [Please distribute the following text to all interested parties]
 [We apologize for any duplicate copies you may receive]

                          CALL FOR PAPERS

         ----------------------------------------------------
         Second International Mozart/Oz Conference (MOZ 2004)
                 Charleroi, Belgium, Oct. 7-8, 2004
                    http://www.cetic.be/moz2004
         ----------------------------------------------------

July 9, 2004:      Paper submission deadline
August 13, 2004:   Notification of acceptance
August 27, 2004:   Camera-ready version due

Information on registration, travel, and accommodation is now
available at the conference Web site http://www.cetic.be/moz2004.
Please note that the deadline for early registration fee is Aug. 15.

The purpose of MOZ 2004 is to bring together researchers, developers,
educators, and users interested in the Oz language and the Mozart
development platform.  Mozart is an advanced platform for intelligent
and distributed applications.  Mozart implements Oz, a multiparadigm
language that combines many concepts in a coherent whole with a simple
semantics and an efficient implementation.  Oz has particular
strengths in concurrent programming, symbolic programming, constraint
programming, and distributed programming.  Mozart and Oz are being
used in research, education, and industry in many areas including
distributed programming, software engineering, programming languages,
constraint programming, simulation, graphical user interfaces,
(Continue reading)

Permalink | Reply | 0 comments |

Gmane