Developing in what image/version/build?

With the plethora of Squeak images, versions, builds, whatever, I'm a
little confused as to picking the best environment.  Currently I've
been working in the 3.9-8 VM with the 3.9-final-7067 image, but some
things don't seem to work right.

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

What's not working?

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

Hi,

        Would it be easier if we work with the StrongTalk VM and embed
primitives that will enable us certification ? If we conclude that the
general VM will have difficulties to pass certification, may be, what we
need is a VM that is certified, which has minimum mods. In which case,
StrongTalk might be a good place to start. Thoughts ?

Cheers
<k/>

_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

It's a very good question.  I'm hoping that the StrongTalk VM will indeed
blossom and become a powerful force in Open Source Smalltalk.  What will
actually happen with it is yet to be seen.  

In the mean time I'm hoping to find that CC validation is possible on
Dynamic Languages like Smalltalk.  I have a call coming up (this Thursday)
with someone that has been through this, but I'm still looking for someone
with experience with a lab or with a validation using a VM language.  

If anyone has some suggestions on someone I should talk to please let me
know.

I'm expecting that we will be able to make the changes necessary to the
current Squeak VM to get it validated.  When StrongTalk is ready maybe we
can do the same, or at least consult on what should be done there.  We still
need a better understanding on what needs to be done.

Ron Teitelbaum
Squeak Cryptography Team Leader


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

On 11/27/06, Ron Teitelbaum <[hidden email]> wrote:

> If anyone has some suggestions on someone I should talk to please let me
> know.

I was reading about Islands over the weekend and was wondering if that
mechanism or something akin to it might be useful.

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

On 11/27/06, Ron Teitelbaum <[hidden email]> wrote:

> What's not working?

For one thing, in some combinations of VM & image Monticello won't see
the Crypto-Config mcz.  In others, MC will see it but it won't load it
(I can try to reproduce when I get home).

Plus I'm just generally confused about the whole
package/changeset/save your project on a server thing.

I've also toyed with the Tweak/iSqueak image (which makes more sense
to me than Morphic, I'm sorry to say) and the Croquet image (which
seems to be having issues with OpenGL initialization on my Mac), but
it's not clear to me how close to the mainline Squeak I should stick.
Or if indeed there *is* a mainline Squeak.  :)

There's a lot to wrap one's mind around when coming from first
principles, and picking a development mode & environment is a big one.

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

Are you talking about croquet islands, or something else?

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

http://minnow.cc.gatech.edu/squeak/2074

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

Hi Tim,

You need to load my patch to get Monticello Configurations to work:
http://bugs.impara.de/view.php?id=5217 

>
> Plus I'm just generally confused about the whole
> package/changeset/save your project on a server thing.

Go signup at www.squeakSource.com I'll add you as a developer.  The right
way to go here is using Monticello.  Getting up to speed is pretty easy.  
http://minnow.cc.gatech.edu/squeak/43 .

The easiest thing is to just save your changes into our repository.  You can
also save you code until you are ready in your package-cache or in a
separate directory.  I run my own squeakSource image to manage my changes.
 
>
> I've also toyed with the Tweak/iSqueak image (which makes more sense
> to me than Morphic, I'm sorry to say) and the Croquet image (which
> seems to be having issues with OpenGL initialization on my Mac), but
> it's not clear to me how close to the mainline Squeak I should stick.
> Or if indeed there *is* a mainline Squeak.  :)

I'm using a 3.9 image.  That's where most of the latest crypto work has been
done.  I suggest you sick with that for now.  

>
> There's a lot to wrap one's mind around when coming from first
> principles, and picking a development mode & environment is a big one.

Yeah, but you are catching on fast!!

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

Interesting I'll have to look more closely at it.

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

Lex was doing this project as an intern in Alan's group back in  
2000.  He recently said somewhere that he launched into the project  
shortly after having been "Millered", a term referring to having  
grokked the writings of Mark S. Miller (see http://erights.org).  The  
word is apt, and makes me chuckle.

I would STRONGLY recommend anyone interested in the area to read  
Mark's thesis dissertation (http://www.erights.org/talks/thesis/).  
It is the most readable dissertation I have ever read, and is  
accessible to the non-specialist (like me).  It is a comprehensive  
and well-structured overview of the field of object-capability  
security.    You won't think the same about security after your first  
Millering.

Josh


On Nov 27, 2006, at 4:43 PM, Ron Teitelbaum wrote:

_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

On 11/27/06, Joshua Gargus <[hidden email]> wrote:
> Lex was doing this project as an intern in Alan's group back in
> 2000.  He recently said somewhere that he launched into the project
> shortly after having been "Millered", a term referring to having
> grokked the writings of Mark S. Miller (see http://erights.org).  The
> word is apt, and makes me chuckle.

Information Assurance is a such a small world.  I had a short debate
in Bruce Schneier's blog earlier this year with Mark Miller re: the
"Petnames" concept in E and how it relates to (and in my mind is
inferior to) Rivest & Ellison's SPKI/SDSI namespaces.

http://www.schneier.com/blog/archives/2006/02/petnames.html

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

I sent an invitation to Lex to join our conversation about islands.

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

There's plenty about 3.9 that is less than acceptable.  I've moved back to 3.8 my forward development until 3.9 improves.

I recommend 3.8 for new Squeakers.  3.9 has some real usability issues that turn off new and old users alike; like lists jumping around upon selection is just one of many.  3.8 offers a cleaner start, less confusing, and you don't have Monticello packages with changes in the base image.  The uninterruptable image lockups though, all on their own, were plenty enough for me to switch back.

3.9 has that new window look, but you can get that in 3.8 real easy if you want:
http://map.squeak.org/package/3f07c64c-e22f-42ef-91e1-40c7e0de9185





_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

That's very interesting, I'm using 3.9 fulltime and I don't have any of the
problems you are experiencing.  Although I'm not using Morphic lists.  How
can I see this bug, do you have a bug number?  I did have the image freezing
problem but we figured that out.  It was the seaside code for comet.
Bumping up the priority fixed the problem.  There were some fixes I had to
do to Monticello but that was only for Configurations.  

I'm sure you very good reasons for using 3.8 I'd like to hear more about
them to see if I can replicate them.  Are you on a Mac, *nix, or Win?

I'm not trying to encourage you to move just curious.  It's good to know
what bugs you can expect to encounter.

Ron



_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

> That's very interesting, I'm using 3.9 fulltime and I don't have any of the
> problems you are experiencing.  

Huh?  Ron, just open a 3.9g-7061 image, what you get from downloading from link at the top of squeak.org, and you have the problems I describe.

Problem #1:  > Monticello packages with changes in the base image

Just open a Monticello browser, see * Morphic and * Multilingual?


Problem #2:  > lists jumping around upon selection

Now just scroll the package list down about 10 and then select a package near the top of the window.  It not jumps straight back to the top, unready for my right-click after I've just left-clicked it.

Lockups happened for me all the time without Seaside; in fact I never was even able to get my few dev tools in there without constant lockups.  I tried to track it down but can't remember now, ran out of time.

>  It was the seaside code for comet.
> Bumping up the priority fixed the problem.

I guess I missed this, can you explain or do you have a link?





_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

See the bug link in:

http://weeklysqueak.wordpress.com/2006/10/26/image-freezing/ 

The problem appears to be task switching on the lowest priority, as long as
your process is not running at the same priority as the idle process
everything appears to work fine.  Also I should note that even though the
image froze up with the comet process seaside was still functional.  So the
image and the process switching were still working but somehow the gui
thread got locked up.  Sometimes accessing seaside would bring back the GUI.
But as I said changing the priority of processes that are running at the
same priority as Idle to at least 1 above idle seemed to fix the problem for
me.

I'm running 7051 but when I get a chance I'll upgrade to the newer image.  

I understand your concern about the open packages.  

I do not have the same list problems that you do.  I open a package pane
browser scroll down select packages and it works great, I don't see any
jumping at all.  I thought maybe you were talking about the Monticello
browser so I opened that and still no problem.  Which browser are you having
trouble with?  I don't see that problem.

I'm using my image all day long 12-14 hours, and I haven't had any image
freeze problems at all.  As I said I did have problems but I figured it out.
I'm running seaside, and the ssl listener, doing all kinds of external
calls.  I'm running Glorp to PostgreSQL, and no freezing.  If you hadn't
said anything I would have said that 3.9 is very stable!  What platform are
you running on and which VM are you using?

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

On 11/27/06, Ron Teitelbaum <[hidden email]> wrote:

> Go signup at www.squeakSource.com I'll add you as a developer.  The right
> way to go here is using Monticello.  Getting up to speed is pretty easy.
> http://minnow.cc.gatech.edu/squeak/43 .

Just FYI, I'm waiting for clearance from my employer.  Since this
stuff is pretty close to what I'm paid to do, I figured better safe
than sorry.  I don't expect a problem from them; this is one of the
advantages of working for a not-for-profit.

-- Tim
_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography

That's a very good idea!

Ron


_______________________________________________
Cryptography mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/cryptography