Is Nobody Using Open Cobalt???

[Richard Kenneth Eng]

This project seems dead. The main website has all kinds of broken links. The code repository is no longer accessible.

Has virtual reality lost its shine, or this is just a Smalltalk/Squeak issue?

[Eric Gillespie]

Hi. I used to chair the Open Cobalt meetings on Skype. At least at the moment, I haven't seen any significant movement on OpenCobalt since I dropped out in early 2011 due to a little earthquake we had, and the weekly Skype meetings haven't happened since then. After that point, it was obvious to me that the principal developers had moved onto other projects that would help keep a roof over their heads. I'm aware there are still original files available now from opencobalt.net (not org), but again, these haven't seen significant movement since then. I'm currently working with another Smalltalk altogether, though not on 3D spaces. I do rather feel a bit like a lone soul waving a small LED torch in the bottom of the Vertical Assembly Building at night.

Sorry to seem the bearer of such bleak news. Also, I haven't seen much activity in the virtual reality sphere aside from game development on various new (since then) head-mounted display devices. I don't think this was limited to Squeak, or even Smalltalk at large.

Regarding the contact information, I'll modify certain pages to let people know there's no longer any meetings. I'm not sure what else I can do aside from that as I lack the skills that both Matthew and John had, so I'm not a good candidate to maintain the project; I suspect as the good Bones would say, "She's dead, Jim".

If any others are willing to correct me on this issue, let me know.

Regards, BrickViking (a.k.a. DrSmokey)

On Sun, 1 Sep 2019 at 13:17, Richard Kenneth Eng <horrido...@gmail.com> wrote:

This project seems dead. The main website has all kinds of broken links. The code repository is no longer accessible.

Has virtual reality lost its shine, or this is just a Smalltalk/Squeak issue?

--
--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/3d8a39e7-3ec5-4cf8-90e6-ac233cfa37e7%40googlegroups.com.

[Aik-Siong Koh]

Hi:

I am sure the project can be revived if there is a good use case to implement.
Are there any ideas out there?

Thanks,
Aik-Siong Koh

Eric Gillespie wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/CAHWye86WAg3mMm5nLPqk%2B2b_LuDSKRx%3DxE2gCmO44xT_UgwaDA%40mail.gmail.com.

[John Dougan]

I was unable to work at all for a while, and when I could again I updated some parts (fixed OBJ importer and redid some of the rendering so it can smooth shade).  But without any interest from outside I switched to other things. Duke is no longer involved, so no resources they hosted (mantis, monticello server, etc.) are available online.  But before the shutdown they dumped a set of archives on me which should have most of it if not everything.

If anyone wants to carry it forward I can get those to you.

There are a number of technical issues in the code base, exactly which I don't recall offhand. I'd have to get my notes. I do have ideas on where it could be taken, and have done a lot of thinking on the problems with VR/VW/Cyberspace/etc. in general (this is like the 4rd or 5th turn of the hype cycle wheel) and what the future will look like. 

If anyone wants to talk about it, I'm up for it.

Cheers,

 -- John

On Sat, Aug 31, 2019 at 6:17 PM Richard Kenneth Eng <horrido...@gmail.com> wrote:

This project seems dead. The main website has all kinds of broken links. The code repository is no longer accessible.

Has virtual reality lost its shine, or this is just a Smalltalk/Squeak issue?

--

--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/3d8a39e7-3ec5-4cf8-90e6-ac233cfa37e7%40googlegroups.com.

--

John Dougan
jdo...@acm.org

[Aik-Siong Koh]

John:

Sure I would like to hear of possibilities for opencobalt. 

Thanks 

Aik-Siong Koh

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/CAJC9MC1uceqq-dcaRaCsEawBxAbyOtkq_rjokrCLDUbUiEqVQQ%40mail.gmail.com.

[Tom]

Hello:  I've long fantasized that I was a member of a citizen's group that presented a VR application to our local city hall.  Installed, it would enable residents to visit city hall, participate in activities ranging from complaints to the mayor or "attending" city council meetings.  "Popping in" to speak with the city clerk seems like a good way to keep local officials on their toes.  Such an application might even be economically sustainable for consultants adding value to a basic application.

Tom Poe, Eden Valley, MN

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/FA564216-3599-4286-B7E2-3185CAD1FB06%40askoh.com.

-- 
4 school shootings in U.S. in one month: 4 school shootings in Germany in 100 years: Laws Matter

[Alan Grimes]

Yeah, much sadness here. I have no idea why Cobalt did not become
mainline Squeak and everything else become legacy...

I have a number of various reasons to want to get this running again,
maybe even get an importer working for popular game asset sets.

Anyway, I would propose to rebase the project on whatever the current
squeak version is and get everything updated, possibly moving to Vulkan
and get everything working and modern so that ppl can jump in.

Lots of work... So step 0 is getting a sponsor of some kind.


--
Clowns feed off of funny money;
Funny money comes from the FED
so NO FED -> NO CLOWNS!!!

Powers are not rights.

[Eugen Leitl]

On Sun, Sep 01, 2019 at 02:39:12PM -0700, John Dougan wrote:
> I was unable to work at all for a while, and when I could again I updated
> some parts (fixed OBJ importer and redid some of the rendering so it can
> smooth shade). But without any interest from outside I switched to other
> things. Duke is no longer involved, so no resources they hosted (mantis,
> monticello server, etc.) are available online. But before the shutdown
> they dumped a set of archives on me which should have most of it if not
> everything.
> If anyone wants to carry it forward I can get those to you.
>
> There are a number of technical issues in the code base, exactly which I
> don't recall offhand. I'd have to get my notes. I do have ideas on where it
> could be taken, and have done a lot of thinking on the problems with
> VR/VW/Cyberspace/etc. in general (this is like the 4rd or 5th turn of the
> hype cycle wheel) and what the future will look like.
>
> If anyone wants to talk about it, I'm up for it.

Hi John -- as a former occasional user of OpenQwaq I was recently
looking at https://www.youtube.com/watch?v=b4nNtN7XBi8 (based on https://github.com/ronsaldo/woden,
discussed in context of immersive VR development on https://news.ycombinator.com/item?id=18267445 )
and was wondering what it would take to make Pharo run OpenCroquet.

I presume it would be a prohibitive amount of work. Am I right? Thanks.

[Richard Kenneth Eng]

I would love to see OpenCobalt running on Pharo. It would make it much easier to promote both Pharo and OpenCobalt. You see, Pharo is the most actively developing Smalltalk today. By comparison, Squeak looks absolutely stagnant.

Pharo has up-to-date books and documentation. Squeak's documentation is outdated, and Squeak tutorials are very disappointing.

Pharo has a wonderful MOOC. Why isn't there a MOOC for Squeak?

Pharo has an industry-supported consortium. There's nothing like that for Squeak.

Pharo is breaking new ground with the Glamorous Toolkit and PolyMath and Roassal. There are Pharo language bindings for TensorFlow. Squeak is AWOL.

I think OpenCobalt would garner more interest if it supported Pharo.

[Eugen Leitl]

On Sun, Sep 01, 2019 at 02:39:12PM -0700, John Dougan wrote:

> I was unable to work at all for a while, and when I could again I updated
> some parts (fixed OBJ importer and redid some of the rendering so it can
> smooth shade). But without any interest from outside I switched to other
> things. Duke is no longer involved, so no resources they hosted (mantis,
> monticello server, etc.) are available online. But before the shutdown
> they dumped a set of archives on me which should have most of it if not
> everything.
> If anyone wants to carry it forward I can get those to you.

I would like to grab whatever you have, though my Smalltalk knowledge is near zero.

I would try to put it up on GitHub/GitLab to make it at least easily accessible.

I was planning to set up a Linux VR workstation by end of this year, specifically looking to play with Pharo.

[Apollo Sande]

Hi all,

If we can have all the available files on a Github/Gitlab repo soon, I am a working on a way we can incentivize it's revival using blockchain-based funding.

Kind regards,

--
--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/20190909120333.GB3513%40leitl.org.

[Richard Kenneth Eng]

Thankfully, Smalltalk is very, very easy to pick up and learn in-depth. I, too, would very much like to see Open Cobalt's revival. VR is the future!

With so much activity in the VR world, I don't understand why Open Cobalt isn't attracting more interest. Surely, Smalltalk isn't the obstacle, is it?

I think Open Cobalt just needs more marketing and advocacy. Go on social media and beat the Open Cobalt drum. Publish a few articles about Open Cobalt.

Putting it on GitLab is a good starting point.

--
--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/20190909120333.GB3513%40leitl.org.

[Eugen Leitl]

On Mon, Sep 09, 2019 at 08:40:45AM -0400, Richard Kenneth Eng wrote:
> Thankfully, Smalltalk is very, very easy to pick up and learn in-depth. I,
> too, would very much like to see Open Cobalt's revival. VR is the future!
>
> With so much activity in the VR world, I don't understand why Open Cobalt
> isn't attracting more interest. Surely, Smalltalk isn't the obstacle, is it?

I've been trying to figure out what happened since the last time I had contact
with a Croquet-derived system (OpenQwak/Teleplace).

Looking at Slack backscroll of Croquet.Studio, here's a nice summary by David A Smith as of 31. July:

"David A Smith 7:00 AM

Just to clarify the versions of Croquet, and why I sometimes refer to the new Croquet as Croquet V.
The first Croquet-like system was ICE, interactive collaboration environment. You can see a video of it here: https://youtu.be/KFdeVRteiSY
The second Croquet-like system was Virtus Open Space. It was the basis of the game engines for Rainbow Six and Timeline. It was a component platform, but unfortunately the collaboration part was lost.

The third Croquet was "classic" Croquet. Written in Smalltalk and created by Alan Kay, David Reed, Andreas Raab and me. It was actually two different versions - pre TeaTime where I developed the concept of a meta message, and simple TeaTime, which was very similar to what we have today in Croquet. You can watch it here: https://youtu.be/XZO7av2ZFB8. This was used as part of Alan Kay's Turing Award lecture that I was incredibly proud to participate in. This was also the platform we used to create Teleplace and lives on in Turf. The fourth version was the Virtual World Framework, which was developed for the DoD while I was a Senior Fellow at Lockheed Martin. This was a complete reboot and was written in JavaScript by David Easter. Here is an amazing application built by Rob Chadwick using VWF - https://youtu.be/utEVly2KCnE. VWF is still being used, but I think of it as more of a prototype to the current implementation. We learned an incredible amount from every one of these implementations and what is so exciting to me is that the current system is not only far smaller and efficient, it is also far more powerful."

So the core devs went on the route to closed source and moving to JavaScript.

More context: https://planetcroquet.squeak.org/

[Richard Kenneth Eng]

> So the core devs went on the route to closed source and moving to JavaScript.

<groan>

I'd rather gnaw off my right leg than program in JavaScript; what a horrid language. For me, it's Smalltalk or nothing!

--
--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/20190909131023.GC3513%40leitl.org.

[Ho-Sheng Hsiao]

On Mon, Sep 9, 2019 at 6:25 AM Richard Kenneth Eng <horrido...@gmail.com> wrote:

> So the core devs went on the route to closed source and moving to JavaScript.

I'd rather gnaw off my right leg than program in JavaScript; what a horrid language. For me, it's Smalltalk or nothing!

I don't like JS much either. However, wasm is gaining a lot of momentum, and maybe that's something to consider.

The Rust community already has a really great tooling targeting it. The Elixir community recently came out with something called Lumen -- targeting Elixir/OTP into the browser, via wasm. (There are warts, particularly with green threads). The Elm folks are planning to move completely to wasm.

I remembered when I tried looking into the OpenCobalt source code, I couldn't use the tools I was used to. (Github; Vim-bindings). It seems to me, from an outsider's perspective, that because Smalltalk is so easy to work with, the community has been working away in isolation from the rest of the open-source community. I can find a lot of information about packages for Ruby, Elixir, Python, but not so much for Smalltalk. None of it gets published there, and it is not discoverable via the web.

-Hosh

 

[Richard Kenneth Eng]

 It seems to me, from an outsider's perspective, that because Smalltalk is so easy to work with, the community has been working away in isolation from the rest of the open-source community. 

It's important to understand that Smalltalk represents a new programming model. Language, (live coding) IDE, and total system of live objects (known as the image) work together in perfect harmony to deliver software quickly and easily. This kind of synergy is unique in the programming world (unless you include Forth and Lisp, both of which are decidedly user-unfriendly).

Live coding practically eliminates the traditional edit-compile-test-debug cycle that has hampered developers for over half a century. It's the main reason that Smalltalk is the most productive general-purpose programming language in the world, according to a study conducted by Namcook Analytics. On average, it's twice as productive as JavaScript, C++, Go, Java, PHP, Python, and C#.

If developers can't let go of the traditional programming model, of course, they will be unhappy. Fortunately, Pharo coexists well with tools like GitHub, so it's not as insular as it used to be.

The question the programming world has to ask itself is, "What is it worth to be able to cut development time in half, or better?" In some instances, the productivity gain can be up to 5X.

--

--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/CAK_OAJV%2Bm_1kYOGi6GtmxiTbJTDuzj%3DLeXR%3DFPO%3DLGEF7UkC6w%40mail.gmail.com.

[Richard Green]

I've a stash of the Smalltalk Sources Croquet code - this is the file header:

'From Croquet1.0beta of 11 April 2006 [latest update: #0] on 11 April 2006 at 3:03 pm'!

[Richard Green]

I also found an slightly newer OpenQwaqV1 sources:

'From Croquet1.0beta of 11 April 2006 [latest update: #1] on 23 February 2009 at 12:02:43 pm'!

OpenQwaq built a python API - which I"ve also stashed.

If folks are interested, I will create repositories on GitHub 

so that others can walk the code and decide if updating it is worth the trouble.

On Saturday, August 31, 2019 at 9:17:34 PM UTC-4, Richard Kenneth Eng wrote:

[Apollo Sande]

Awesome, please repo it.

--

--
You received this message because you are subscribed to the "Open Cobalt" Google group.
To post to this group, send email to openc...@googlegroups.com
To unsubscribe from this group, send email to
opencobalt+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/opencobalt?hl=en

---
You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/e7e046da-3166-4e42-af24-d38ab0c8db39%40googlegroups.com.

[Alan Grimes]

It needs to be pointed out that Squeak is its own private universe of
tools and conventions.

If you don't have a Squeak 5.2 image, go get one...

When you fire it up, go to the "tools" menu item, there will be a group
of entries called "montechello"

"Montechello" is smalltalk for Git.

I can't fathom why such an obscure "codeword" is presented to the user
like this but there's nothing I can do about it.

Anyway, Montechello is the package repository system, it is also how you
update installed packages though you have to hunt around for the right
popup menu that will give you the option of checking all packages for
updates.


Also, initial vulkan report.

1. Squeak does have good XML tools.
2. The vulkan schema is described at:
https://www.khronos.org/registry/vulkan/specs/1.1/registry.html

3. status of FFI in Squeak 5.2, unknown -- requires VM support, not sure
what the current status is.

4. Vulkan is fundamentally a framework for running shader programs
written in OpenCL, and SPIR bytecode on GPU hardware. -- so in addition
to writing an interface for Vulkan proper, a system for dealing with
shader programs would also be needed.

Apollo Sande wrote:
> Awesome, please repo it.
>
>
> On Fri, 13 Sep 2019, 00:11 Richard Green,
> <richardale...@gmail.com

> <mailto:richardale...@gmail.com>> wrote:
>
> I also found an slightly newer OpenQwaqV1 sources:
> 'From Croquet1.0beta of 11 April 2006 [latest update: #1] on 23
> February 2009 at 12:02:43 pm'!
> OpenQwaq built a python API - which I"ve also stashed.
> If folks are interested, I will create repositories on GitHub
> so that others can walk the code and decide if updating it is
> worth the trouble.
>

[Eugen Leitl]

On Fri, Sep 13, 2019 at 02:30:38AM +0300, Apollo Sande wrote:
> Awesome, please repo it.

This isn't the latest OpenQwaq repo then https://github.com/itsmeront/openqwaq ?

Latest commit
c226fee
on Feb 5, 2012

What about potential license impedance mismatch?
I thought Open Cobalt was MIT only, while OpenQwaq is GPL v2.0.
You would have to clean-room backport things from there.

Notice that Nikolai Suslov is working on reviving Open Croquet in the current Squeak http://files.squeak.org/various_images/OpenCroquet

https://github.com/NikolaySuslov/croquet-squeak

https://sdk.krestianstvo.org/sdk/croquet.html/

Relevant discussion threads on Smalltalk forums:

http://forum.world.st/Open-Croquet-will-ahve-issues-in-next-version-of-Mac-OS-X-td5087553.html

http://forum.world.st/ANN-Open-Croquet-for-Squeak-5-2-32bit-release-td5087739.html

> > <https://groups.google.com/d/msgid/opencobalt/e7e046da-3166-4e42-af24-d38ab0c8db39%40googlegroups.com?utm_medium=email&utm_source=footer>
> > .

> >
>
> --
> --
> You received this message because you are subscribed to the "Open Cobalt" Google group.
> To post to this group, send email to openc...@googlegroups.com
> To unsubscribe from this group, send email to
> opencobalt+...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/opencobalt?hl=en
>
> ---
> You received this message because you are subscribed to the Google Groups "Open Cobalt" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to opencobalt+...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/opencobalt/CAEMHUQAOhsdhh0tvOkV%2B0DmtK9AOxGfUfgH1Umg5FurfpQvuKw%40mail.gmail.com.