RealXtend Server 0.62 - Was: Re: [REX] Re: What is the status of ModRex?

[Mark Malewski]

Also, would it be possible to begin using RealXtend server release numbers, that are in-line with the current OpenSim release numbers?  For example create a RealXtend Server 0.62, that is using the current/latest OpenSim 0.62 Server build, and using the latest ModRex build, and also please include the database from the RealXtend 0.4 Beneath the Sea database build?  This way the realXtend Server release numbers are in-line with the OpenSim Server release numbers?

 

This way users could just download one single server file from the RealXtend website, install it and be running the latest OpenSim 0.62 release with the latest ModRex, and have the "Beneath the Sea" database already up and running?

As bugs are fixed, we could always have different build numbers for example RealXtend 0.62 - Build ####, as necessary, this way as bugs are fixed we could post updated build numbers to the realXtend site.

 

This way we could keep the realXtend 0.62 Server "all in one" download setup a very simple and easy to use setup that will get the latest OpenSim 0.62 release installed with the latest ModRex, and the "Under the Sea" database all running without users having to do their own binary compiles or builds?

 

                  Mark

 

On Thu, Feb 12, 2009 at 9:51 PM, Mark Malewski <mark.m...@gmail.com> wrote:

Could someone post a compiled binary of the latest Modrex to the realXtend website for Windows?

 

Like just compile the current ModRex binary, and call it "ModRex 0.62 - Build ####" (in relation to the current OpenSimulator 0.62 current release).

 

This way people could just download the latest ModRex download/zip/binary and get up and running fairly quickly?

 

               Thanks,

 

                 Mark

On Thu, Feb 12, 2009 at 2:16 PM, Gustavo Alberto Navarro Bilbao <alberto.nav...@gmail.com> wrote:

Hello Mark:

You can find a guide to compile Modrex here:
http://opensimulator.org/wiki/ModRex

I'm tryng to make a test this

Good Luck/Suerte ;-)

========================================================

2009/2/12 Mark Malewski <mark.m...@gmail.com>:

> What is the Status of ModRex, and also how do I run ModRex?
>
> I was previously using RealXtend Server 0.4, but I've downloaded the latest
> OpenSim 6.2 build and I'm just trying to figure out what the difference
> between OpenSim 6.2 is, and how do I get OpenSim 6.2 working with RealXtend?
>
> What exactly is ModRex?  Is that a module, and if so where can I download
> it, and how do I install it in OpenSim 6.2?
>
> Are there any tutorials on how to use ModRex with OpenSim 6.2?  Also, will
> ModRex give me the ability to use the current RealXtend standalone Avatar
> Storage, and the standalone Authentication Server?
>
>                 Thank-you,
>
>                     Mark
> >
>

[Mikko Pallari]

ModRex is still far from ready, as its still missing some core features. We surely inform this mailing list, when we are ready for wider testing.

 

______________________________

Mikko Pallari

Server Technical Lead in RealXtend

 

ADMINO technologies

www.adminotech.com

www.realxtend.org

[Ryan McDougall]

We are definitely interested in clarifying release numbering and
compatibility with OpenSim. However our work is still very much
on-going, and ModreX is simply not yet ready for general release.

One of the problems with making an all in one at this time is the
ModreX has regressions compared to the old forked version of OpenSim
used in 0.4 that would require many awkward hacks to replicate in the
cleaner ModreX. It was felt that with reX-NG viewer coming it would be
better to focus on keeping ModreX clean, and fixing the underlying
problems instead of making hacks for the sake of backwards
compatibility. That means that the "Beneath the Sea" in 0.4 will
likely never run on ModreX unmodified.

Please keep in mind the very alpha nature of realXtend at this point.
We are doing our best to improve the quality, but like any renovation,
it requires a period of demolition.

Cheers,

[Peter Quirk]

A couple of questions about modrex:

1. Is there a list of areas known to be not implemented? Several of us
are filing bug reports against modrex without knowing whether
we are investigating known areas of unimplemented functionality.

2. Is there a target release date?

Thanks,
-- Peter

> > On Thu, Feb 12, 2009 at 9:51 PM, Mark Malewski <mark.malew...@gmail.com>

> > wrote:
>
> >> Could someone post a compiled binary of the latest Modrex to the realXtend
> >> website for Windows?
>
> >> Like just compile the current ModRex binary, and call it "ModRex 0.62 -
> >> Build ####" (in relation to the current OpenSimulator 0.62 current release).
>
> >> This way people could just download the latest ModRex download/zip/binary
> >> and get up and running fairly quickly?
>
> >>                Thanks,
>
> >>                  Mark
>
> >> On Thu, Feb 12, 2009 at 2:16 PM, Gustavo Alberto Navarro Bilbao
> >> <alberto.navarro.bil...@gmail.com> wrote:
>
> >>> Hello Mark:
>
> >>> You can find a guide to compile Modrex here:
> >>>http://opensimulator.org/wiki/ModRex
>
> >>> I'm tryng to make a test this
>
> >>> Good Luck/Suerte ;-)
>
> >>> ========================================================
>

> >>> 2009/2/12 Mark Malewski <mark.malew...@gmail.com>:

> >>> > What is the Status of ModRex, and also how do I run ModRex?
>
> >>> > I was previously using RealXtend Server 0.4, but I've downloaded the
> >>> > latest
> >>> > OpenSim 6.2 build and I'm just trying to figure out what the difference
> >>> > between OpenSim 6.2 is, and how do I get OpenSim 6.2 working with
> >>> > RealXtend?
>
> >>> > What exactly is ModRex?  Is that a module, and if so where can I
> >>> > download
> >>> > it, and how do I install it in OpenSim 6.2?
>
> >>> > Are there any tutorials on how to use ModRex with OpenSim 6.2?  Also,
> >>> > will
> >>> > ModRex give me the ability to use the current RealXtend standalone
> >>> > Avatar
> >>> > Storage, and the standalone Authentication Server?
>
> >>> >                 Thank-you,
>

> >>> >                     Mark- Hide quoted text -
>
> - Show quoted text -

[Mikko Pallari]

Hi,

This is somewhat accurate list of features supported currently: http://rexdeveloper.org/wiki/index.php?title=Features_supported_currently

And here is a list of python methods and events currenly supported: http://rexdeveloper.org/wiki/index.php?title=PythonInModRex

Cheers,
Mikko

[Mark Malewski]

Ryan/Peter,

 

> compatibility. That means that the "Beneath the Sea" in 0.4 will
> likely never run on ModreX unmodified.

Sounds like an interesting project.  So it would be possible to "modify" it to work with RexMod?  I might give this one a shot though.  Just because I was really impressed by the hard work that the RealXtend/Ludocraft crew did on "Beneath the Sea".  I may try to give it a shot, and see if I can get the current "Beneath the Sea" working with the new ModRex.  It would be a good "learning experience" for me.  How hard would it be to modify "Beneath the Sea" to work with ModRex?  I hate to see "Beneath the Sea" get wasted, so maybe I'll work on updating it for the new ModRex.  (I may need a bit of a hand with this project, if anyone is good with building/coding/scripting).

 

> We are doing our best to improve the quality, but like any renovation,
> it requires a period of demolition.

Yes, I understand.  I certainly do look forward to the new RealXtend browser (reX-NG Viewer).  Now the new reX-NG viewer will remain 100% compatible with OpenSim, and ModRex correct?

 

>1. Is there a list of areas known to be not implemented? Several of us
>are filing bug reports against modrex without knowing whether
>we are investigating known areas of unimplemented functionality.

Yes, this is a very good question.  It would probably be good to continue filing the bug reports, just so the developers are aware of the problems (and may be able to eventually post some "work arounds" for the incompatibility issues and at least all the areas of "unimplemented functionality" are documented).

 

> 2. Is there a target release date?

Another very good question.  I hate to bother asking, but yes is there any type of scheduled "target" release date for ModRex?  Just so we have an idea?  Also what Viewer do you suggest we use in the interim with the new ModRex?  Will the current 0.4 RealXtend Viewer work with ModRex?  Does Hippo work with ModRex?  Also will the new Viewer by Kirsten Lee (r16? or r17?) work with the new ModRex?  Any ideas on an expected release date for that new viewer from Kirsten Lee? 

 

              Mark

[Mark Malewski]

Mikko,

 

Excellent information.  Thank-you for those updated Wiki links, I'll bookmark them.  It at least helps us see what is coming, and the status.  Thank-you!

 

          Mark

[Lc]

i made this tuts in order to recreate some part of "under the sea"
http://www.rexdeveloper.org/wiki/index.php?title=Tutorial_%22the_Atoll%22_based_on_%22under_the_sea%22

As far as i know, hte volumedetection won't work atm

Sacha

[Mark Malewski]

LC,

 

Looks good, thanks for the tutorial!  I may use it as a "starting point" and maybe we can slowly work together on updating the "Beneath the Sea" region for ModRex.  I really liked the demo that RealXtend did for "Beneath the Sea", and it would be good to go back and remake the "Beneath the Sea" demo for ModRex.  This way people can use it as a starting point/demo for learning.

 

Eventually I'd like to eventually do several demos (an airport, a naval base, a railroad, etc.).  Then make the demos available for download.  This way people can download one of the ReX demo's, and see how we did the building for a particular region (with cars, aircraft, ships, vehicles, etc.).

 

             Mark

[Jeroen van Veen]

any idea if save-oar and load-oar work for rexobjects?

On Monday 16 February 2009 14:35:57 Mark Malewski wrote:
> LC,
>
> Looks good, thanks for the tutorial! I may use it as a "starting point"
> and maybe we can slowly work together on updating the "Beneath the Sea"
> region for ModRex. I really liked the demo that RealXtend did for "Beneath
> the Sea", and it would be good to go back and remake the "Beneath the Sea"
> demo for ModRex. This way people can use it as a starting point/demo for
> learning.
>
> Eventually I'd like to eventually do several demos (an airport, a naval
> base, a railroad, etc.). Then make the demos available for download. This
> way people can download one of the ReX demo's, and see how we did the
> building for a particular region (with cars, aircraft, ships, vehicles,
> etc.).
>
> Mark
>
> On Mon, Feb 16, 2009 at 8:16 AM, Lc <lcc...@gmail.com> wrote:
> > i made this tuts in order to recreate some part of "under the sea"
> >
> > http://www.rexdeveloper.org/wiki/index.php?title=Tutorial_%22the_Atoll%22
> >_based_on_%22under_the_sea%22
> >
> > As far as i know, hte volumedetection won't work atm
> >
> > Sacha
> >
> > On Mon, Feb 16, 2009 at 3:09 PM, Mark Malewski
<mark.m...@gmail.com>wrote:
> >> Ryan/Peter,
> >>

> >> *> compatibility. That means that the "Beneath the Sea" in 0.4 will

> >>
> >> > likely never run on ModreX unmodified.
> >>

> >> *

> >> Sounds like an interesting project. So it would be possible to "modify"
> >> it to work with RexMod? I might give this one a shot though. Just
> >> because I was really impressed by the hard work that the
> >> RealXtend/Ludocraft crew did on "Beneath the Sea". I may try to give it
> >> a shot, and see if I can get the current "Beneath the Sea" working with
> >> the new ModRex. It would be a good "learning experience" for me. How
> >> hard would it be to modify "Beneath the Sea" to work with ModRex? I
> >> hate to see "Beneath the Sea" get wasted, so maybe I'll work on updating
> >> it for the new ModRex. (I may need a bit of a hand with this project,
> >> if anyone is good with building/coding/scripting).
> >>

> >> *> We are doing our best to improve the quality, but like any

> >> renovation,
> >>
> >> > it requires a period of demolition.
> >>

> >> *

> >> Yes, I understand. I certainly do look forward to the new RealXtend
> >> browser (reX-NG Viewer). Now the new reX-NG viewer will remain 100%
> >> compatible with OpenSim, and ModRex correct?
> >>

> >> *>1. Is there a list of areas known to be not implemented? Several of us

> >>
> >> >are filing bug reports against modrex without knowing whether
> >> >we are investigating known areas of unimplemented functionality.
> >>

> >> *

> >> Yes, this is a very good question. It would probably be good to
> >> continue filing the bug reports, just so the developers are aware of the
> >> problems (and may be able to eventually post some "work arounds" for the
> >> incompatibility issues and at least all the areas of "unimplemented
> >> functionality" are documented).
> >>

> >> *> 2. Is there a target release date?
> >> *

[Mark Malewski]

Jeroen,

 

> any idea if save-oar and load-oar work for rexobjects?

Another very good question.  I was actually going to try that out this week with "Beneath the Sea".  If anyone knows the answer, please post.  I was going to try and backup it up and then just restore it to a clean region from within OpenSim 6.2, but I just wasn't sure what would even happen.

 

Can anyone shed some light on this?

               Mark

[Ryan McDougall]

On Mon, Feb 16, 2009 at 4:09 PM, Mark Malewski <mark.m...@gmail.com> wrote:
> Ryan/Peter,
>
>> compatibility. That means that the "Beneath the Sea" in 0.4 will
>> likely never run on ModreX unmodified.
>
> Sounds like an interesting project. So it would be possible to "modify" it
> to work with RexMod? I might give this one a shot though. Just because I
> was really impressed by the hard work that the RealXtend/Ludocraft crew did
> on "Beneath the Sea". I may try to give it a shot, and see if I can get the
> current "Beneath the Sea" working with the new ModRex. It would be a good
> "learning experience" for me. How hard would it be to modify "Beneath the
> Sea" to work with ModRex? I hate to see "Beneath the Sea" get wasted, so
> maybe I'll work on updating it for the new ModRex. (I may need a bit of a
> hand with this project, if anyone is good with building/coding/scripting).

I think it would be an interesting project to modify it to work.
However, I suspect that the modified version would end up being fairly
different. At least the attempt would tell us more about all
regressions.

>> We are doing our best to improve the quality, but like any renovation,
>> it requires a period of demolition.
> Yes, I understand. I certainly do look forward to the new RealXtend browser
> (reX-NG Viewer). Now the new reX-NG viewer will remain 100% compatible with
> OpenSim, and ModRex correct?
>
>>1. Is there a list of areas known to be not implemented? Several of us
>>are filing bug reports against modrex without knowing whether
>>we are investigating known areas of unimplemented functionality.
> Yes, this is a very good question. It would probably be good to continue
> filing the bug reports, just so the developers are aware of the problems
> (and may be able to eventually post some "work arounds" for the
> incompatibility issues and at least all the areas of "unimplemented
> functionality" are documented).

See the link Mikko posted. If you have information to add, please do,
but clear it with Mikko first.

>> 2. Is there a target release date?
> Another very good question. I hate to bother asking, but yes is there any
> type of scheduled "target" release date for ModRex? Just so we have an
> idea? Also what Viewer do you suggest we use in the interim with the new
> ModRex? Will the current 0.4 RealXtend Viewer work with ModRex? Does Hippo
> work with ModRex? Also will the new Viewer by Kirsten Lee (r16? or r17?)
> work with the new ModRex? Any ideas on an expected release date for that
> new viewer from Kirsten Lee?

There is no current release plan for ModreX except to get as much of
it done as soon as possible. It is the server component we will be
developing the viewer component against, so it needs to be done for
reX-NG. Probably the first "release" of it will be as a whole as
"reX-NG".

> Mark

Cheers,

[Lc]

Hi all

Quote:

There is no current release plan for ModreX except to get as much of
it done as soon as possible. It is the server component we will be
developing the viewer component against, so it needs to be done for
reX-NG. Probably the first "release" of it will be as a whole as
"reX-NG".

I disagree with that. Some of us want to test modrex in our grid. We need at least a regular working modrex from time to time in order to create, test contents.
Modrex without contents will be a failure and according the high technicity behind Rex, it won't be easy to bring creators working on it.

Providing a binary will more than enough at the moment.

Thanks to think about that.
Sacha

[Antti Ilomäki]

Greetings

Like Ryan said nothing has been set in stone about modreX except that
we're working hard to get it done as soon as possible. There are some
open questions regarding for example possible changes in the present
viewer and when we know more about our plans in the near future we'll
also have a release schedule for modreX. This should be quite soon, I
hope.

2009/2/17 Lc <lcc...@gmail.com>:

[Ryan McDougall]

I think there is nothing wrong with community contributed binaries
based on what is publicly available.

Please understand that with our extremely limited resources, any time
spent making ModreX work on legacy demos means we are not spending
time on the future.

That is not to say no one can make ModreX work where you want it to,
it just means the community has to step in a contribute what they
need. Scratch their own itch as it is.

If there is something you need from reX, and you want to do it
yourself, let me know and I'll do what it takes to enable you to help
yourself.

Cheers,

[Lc]

i gave up

[Lc]

we are not talking about legacy, we are talking about creating contents. Creating contents cost money too.
Sm

[Antti Ilomäki]

Content creation is very important (and expensive to us). It is also
one of the most important considerations for us when we decide the
tasks we need to get done next. Having people work on realXtend,
hosting their own worlds and creating content for each other to enjoy
is vital to us, because we don't intend to host large grids ourselves.

The fact is we could always use help from the community, but we also
realize that especially on the SL-viewer based software it could be an
issue. I will know more in the near future, but right now I just don't
have enough information available to give realistic dates and content
of our next releases. We are currently in a planning & design -phase
that is scheduled to last until the end of this month, at that time I
should have more information to hand out.



2009/2/17 Lc <lcc...@gmail.com>:

[Mark Malewski]

Wow, this is getting ugly.

 

Ok, I've been reading "both sides" of this, and I can see how Developers want to wait till everything is 100% done, before letting anyone use it.

 

Well, the problem is... just like LC has said, the "early adopters" need to at least begin using it so that tutorials can be written, and that we can begin to lay the "ground work" for others to follow.

 

I understand developers like Mikko are working as fast as humanly possible, and I thank him for all his hard work, but at the same time we need people like LC and myself who can begin using some of this stuff (so we can begin experiment/testing/documenting).  The documentation needs to be done hand-in-hand with the development.

 

Without any documentation, there is no finished product.  Users need documentation.  They need something they can read, so they UNDERSTAND how to use it.  I've been sifting through what we have, but the documentation really needs to be improved quite a bit, and we need more tutorials.  (Similar to "Beneat the Sea") so that users can download these worlds/regions, and begin to see WORKING EXAMPLES of how RealXtend/ModRex works (and see all the "latest and greatest features" in action).

 

Plus it gives "newbies" an opportunity to pick apart the code, and learn from it.

 

> I disagree with that. Some of us want to test modrex in our grid. We need at least a regular working modrex from time to time in order to create, test contents.

I agree with Sacha.  It would be similar to saying "Let's wait will OpenSim hits a 1.0 release and is finished before we begin to use it or test it."

 

I understand ModRex isn't completely finished, but we need a working binary, and we need to begin using it (for testing purposes) so we can at least give the developer's some feedback, and also begin to work on documentation and testing.

 

Releasing a new binary at least once every 2-3 weeks (in the early "Alpha" stages would help us tremendously in bringing ModRex to Beta).

At least we can give feedback as to what is working, and what isn't working.  At least we can begin testing the 85% of features that have already been implemented in ModRex, and at least we can give some feedback as to whether they are working or not.

 

We'll NEVER get to "green" in the progress chart, if we don't start working on the testing, implementing, and documentation of these ModRex features and commands.

 

> We need at least a regular working modrex from time to time in order to create, test contents.
Modrex without contents will be a failure and according the high technicity behind Rex, it won't be easy to bring creators working on it.

 

I completely agree.  The content developers need something to begin experimenting, and testing, and a working binary that is updated/compiled every 2-3 weeks would really help out a LOT.

 

> Providing a binary will more than enough at the moment.

If I need to, I'll begin working on this (if no one else will).  But I'm just up to my eyeballs in projects right now.  It would be extremely helpful if we could get someone to post compiled binaries to the sourceforge, or the RealXtend site.  Just so we can begin testing ModRex.

 

                     Mark

[Mark Malewski]

Ok, I'm slowly "catching up" on reading these posts.  In response to what Ryan has said:

 

> with our extremely limited resources, any time
> spent making ModreX work on legacy demos means we are not spending
> time on the future.

 

I understand the ModRex developers are working hard on getting ModRex working (and functional).  This is important. 

 

What I'm saying is that the "community" needs at least a ModRex Alpha release, so the grid owners can at least post a "ModRex Alpha Build" so that content creators can begin working on developing ModRex content (and writing documentation and tutorials).

 

>That is not to say no one can make ModreX work where you want it to,
>it just means the community has to step in a contribute what they
>need. Scratch their own itch as it is.

 

Exactly correct.  I understand the "community" of developers need to step up, and I guess if the resources are "limited" then I may have to take the time to work on doing binary compiles.  Just so others can at least use them.  (Just remind me offline, every 2-3 weeks to do a new compile).  Just so the Alpha Testers will at least have something to work with (just don't cry to me if it's broken, or not stable). 

 

> If there is something you need from reX, and you want to do it
> yourself, let me know and I'll do what it takes to enable you to help
> yourself.

Fine, I've compiled the Viewer, I'll work on compiling ModRex.  I just have a lot on my plate at the moment, but once I get a compile done, then I'll ask for access to SourceForge, or at least ask for a place that I can upload the binaries, and update the links so that I can post the latest Alpha binaries so people can begin testing ModRex.

 

Nothing will get done this week, but I'll try to have something done by Monday.  I'll get cracking over the weekend, and if Ryan can get me some access to an FTP site, or at least get me in contact with whoever is hosting the RealXtend website/web page, and I'll work on posting/updating links, and getting the latest ModRex binaries posted about every 2-3 weeks.

 

This would take a load off of RealXtend developers, but at the same time it will get the RealXtend content developers (and users) working on using RealXtend ModRex (which I believe is the KEY to moving this project forward).

 

I understand a new Viewer is going to be another big step as well.  I may see if I can work with the Hippo crew, and see if I can at least get ModRex functionality implemented into the latest release of Hippo, and maybe just work on a new Hippo build myself (if I have to).  Just so we will have some ModRex functionality in the "Interim" while RealXtend developers are working on a new "Rex-NG" Viewer.

 

This way the RealXtend crew can focus all their energy and efforts on development, and at least the RealXtend community will have something to work with in the Interim.  I believe this would probably be the best "logical step" forward.

 

I'll talk to Ryan about getting some FTP access, and maybe getting access to the Downloads page on the RealXtend website, and I'll work on getting some binaries compiled this week.

 

             Thanks,

 

              Mark

 

Rome wasn't built in a day...  ;-)

[Mark Malewski]

LC,

 

>  i gave up

 

Lc, don't give up.  Relax.  We'll get this done, and make it happen.

 

I'll just work with Ryan, and I'll get the binaries compiled myself, and get them posted for others.

 

I'll try to have something up/done by Monday.  I'm just up to my eyeballs in projects right now, but yes I can see how this is important.  So I'll get working on this, and I'll contact Ryan offline about getting some FTP access, and getting access to the downloads page so I can at least get some links posted for the new ModRex Alpha builds.

 

> Content creation is very important (and expensive to us). It is also
> one of the most important considerations for us when we decide the
> tasks we need to get done next. Having people work on realXtend,
> hosting their own worlds and creating content for each other to enjoy
> is vital to us, because we don't intend to host large grids ourselves.

 

Yes, content creation is extremely expensive.  I'll work with LC on this, and we'll at least try to help with some content creation for the new ModRex.  One of my personal tasks will be to try and redo the "Beneath the Sea" demo, so that it will work with RealXtend ModRex.  I believe it would be a good contribution to the community, and then later  maybe we can begin will something more advanced (like Airports, Naval Base, and a Railroad, cars, etc.)

 

I'll host a grid.  Just so we have a "sandbox" to play in for developing for ModRex/RealXtend.

 

We'll need to setup a "RealXtend University".  Once I get the grid up, then I'll put a call out to developers.  I'll probably work hand-in-hand with LC on this.  I have tons of server equipment laying around (Servers with dual 8-Core XEON processors, 128GB ram, etc.) and maybe I'll just setup a grid so we can do some testing/development with RealXtend ModRex.

 

I'll try to work with LC on this stuff.  Hopefully we can get some people that are good in graphics design to come join the Grid, and at least begin working/developing on demos for the new ModRex RealXtend.  This way we can package these "demo worlds" up, as part of a new ModRex release (once everything is finished with ModRex and once we get out of Alpha stage, and get into a Beta stage).

 

> The fact is we could always use help from the community, but we also
> realize that especially on the SL-viewer based software it could be an
> issue.

 

Yes, this is/could be an issue.  I will see if I can possibly get working with the Hippo crew, and maybe release a new "ModRex" compatible Hippo Viewer.  Just so we have something working/available for the community to use.  This will at least give the RealXtend crew time to work on their new "Rex-NG" Viewer, and at the same time users can at least begin building and using ModRex.

 

> We are currently in a planning & design -phase
> that is scheduled to last until the end of this month, at that time I
> should have more information to hand out.

 

Ok, well thank-you.  Just continue your work on developing ModRex and the Rex-NG Viewer, and I'll try to help/work on getting the binaries released, and posted.  I'll try to get some help from LC and we'll also try to work on documentation, tutorials, and content building for the new ModRex.  This way we are working together in unison, and hopefully we'll have some Demos available when ModRex leaves Alpha, and begins to hit Beta.

 

At least this way we can begin reporting any bugs for ModRex, and also begin making "feature requests". 

 

                Mark

[Lc]

i won't give up, but actually it's really frustrating...

[Ryan McDougall]

On Tue, Feb 17, 2009 at 6:14 PM, Mark Malewski <mark.m...@gmail.com> wrote:

>
> Ok, I've been reading "both sides" of this, and I can see how Developers
> want to wait till everything is 100% done, before letting anyone use it.
>
> Well, the problem is... just like LC has said, the "early adopters" need to
> at least begin using it so that tutorials can be written, and that we can
> begin to lay the "ground work" for others to follow.

There may be a lot of room for misunderstanding in this thread, but
let me try and make a few points.

I don't think there is any case of wanting "100%" done. I have been
working on open source for my entire professional career, and I
understand "release early, release often". However I also understand
that at this point reX is alpha software, and alpha software isn't
meant for users yet. It's meant for people with the stomach for
downloading from SVN, compiling it themselves, and helping with bug
reports. The source is freely available, and building it isn't very
difficult compared to any other project in an alpha status.

What I understand you're asking is:

1. When ModreX will become beta, and thus one might presume be easier
to build, and worth making a binary release of?

The answer is when its roughly feature complete (not to be confused
with 100% complete in every way). That means its not missing huge
gaping holes of functionality.

The reason why is simple: because until we have patched the all the
huge gaping holes in features, we are too busy doing fundamental work
of the first order, before we can work on polishing it ready for beta
user consumption. Its purely a matter of limited resources.

If, however, you intend to contribute resources yourself, that
equation changes. Give me a precise list of what you need to start
with.

2. Will ModreX support "Under the Sea" demo completely?

The answer is unlikely, just yet.

The reason is that the previous versions of software contained
work-arounds for SL et al., we would rather make things work properly
in the first place. That means that for a period of time (the duration
of which is unknown), there *will* be regressions due to renovation.
Precisely where those regressions are, and when they might be fixed,
is only partly known. What is know is that regressions are a fact of
software engineering, and pretending they don't exist is foolhardy.

However, if you intend to contribute your own resources to debugging
"Under the Sea", and generate a list of regressions for the community,
then we would eventually know where they all are, and we would be one
step closer to having them fixed.

Bottom line, this is a community of open source users and developers.
It only works when everyone is lending their hand. Otherwise, I am
afraid you'll have to wait for our hands to become freer.

Lastly, if the only issue here is getting a compiled binary in your
hands, I am sure Mikko can find his latest DLL somewhere. However,
don't expect it to be smooth sailing. That'd be like expecting an egg
to come out of the fridge boiled.

Cheers,

[doeko....@gmail.com]

On 17 feb, 19:10, Ryan McDougall <sempu...@gmail.com> wrote:

> On Tue, Feb 17, 2009 at 6:14 PM, Mark Malewski <mark.malew...@gmail.com> wrote:
> I don't think there is any case of wanting "100%" done. I have been
> working on open source for my entire professional career, and I
> understand "release early, release often". However I also understand
> that at this point reX is alpha software, and alpha software isn't
> meant for users yet. It's meant for people with the stomach for
> downloading from SVN, compiling it themselves, and helping with bug
> reports. The source is freely available, and building it isn't very
> difficult compared to any other project in an alpha status.

I see where you are coming from, and I partially agree. However, there
is an amount of users that, I think, has enough technical
understanding to set things up and could contribute bug reports.
However, they don't have the technical knowhow to do a full compile,
or simply not the amount of time required to figure out how to do it
due to lack of experience. I know I am one of those people.

I do not quite see what the problem is with just releasing an alpha
binary with a huge warning sign "HEY, THIS IS NOT FINISHED!". Same way
opensim does it.

The way I understand the status of ModRex is currently, is that it
lacks a number of features that it needs to be considered "featureful
enough" by the REX team to be ocnsidered beta. However, in my opinion,
the stuff that needs to work for it to be worthwhile to release for
many people, in my opinion is simply:
1) not crash every 2 minutes
2) accept .mesh files (and collission meshes etc).
(optional) 3) hopefully be compatible with future modrex releases

It already does this and far more. So, why not release a preliminary
binary for this community of "pro users" (as I'll call it...) to help
out.

As long as you keep the download info fairly obscure (i.e. not put it
on the download page but make it clear to everyone who looks for it
where or from whom to get it), I do not see how this could harm
RealXtend.

Thanks, I hope you understand my point of view.

[Mark Malewski]

 

> As long as you keep the download info fairly obscure (i.e. not put it
> on the download page but make it clear to everyone who looks for it
> where or from whom to get it), I do not see how this could harm
> RealXtend.

 

This is another VERY good point.

 

Too many users are coming here looking for "end user" support (hand holding) and at this point, we're still in very early Alpha stages.  So there are a handful of people here that are willing to be "Alpha Testers" (and stomach the pains of Alpha testing), but I do agree that compiling a binary is a bit of a pain.

 

I do have a machine setup specifically for compiling, but I just haven't gotten around to downloading and compiling the latest SVN.  I can do it, but just haven't taken the time to do it (have a million other projects going at the moment).  It's not that I can't do it, but it would take a developer about an hour or two to just run the build (while they are eating lunch), and just post the latest binary every 2-3 weeks. 

 

If it comes down to "resources" then I guess I'll just have to stomach the task, and "take one for the team" and just do it myself, and post it for the Alpha Testers.

 

Yes, now that Doeko mentions it, maybe posting to the RealXtend website might not be a good idea.  There are too many "new users" that might stumble upon it, and then it would unleash a wrath of headaches (as they complain that it doesn't work, no documentation, etc.)

 

So maybe I will just run the compiles, and host the binaries on a separate page (just for Alpha testing).  This way I can post a link in the forums, and maybe we can "offload" some of the headaches from the Developers.

 

This way the Alpha Testers can work together, and we can help file bug reports, and have intelligent conversation about the problems we are facing.  Then we can bring these problems back to the main development team, and see what they have to say about it.

 

>So, why not release a preliminary
>binary for this community of "pro users" (as I'll call it...) to help
>out.

 

I believe this is what LC, you (and even myself) have all been thinking.

 

> As long as you keep the download info fairly obscure (i.e. not put it
> on the download page but make it clear to everyone who looks for it
> where or from whom to get it), I do not see how this could harm
> RealXtend.

 

I'm going to get the source compiled, and then I'll post the binaries.  I'll try to get a page setup on one of my servers, and just post the latest binaries for Alpha Team testers.  I'll probably setup a separate forum for Alpha Testers, and this way we are not blowing up the message lists, and the development team can join the Alpha Team mailing list as well (if they want).  This way we can discuss Alpha Team testing (setup, configurations, problems with latest builds, etc.) while not blowing up this list, or bombarding everyone else with this discussion.

 

Yes, the mainstream will need to wait (for a stable Beta build), but the Alpha Team can at least get in early, and just download the binaries, and begin having intelligent discussion about the problems, bugs, and issues that we are experiencing so that we can give good feedback to the developers.  This will help them progress, and slowly we'll get closer to a Beta.

 

I know the realXtend crew is just flooded at the moment, so I'll try to "Offload" a bit of the Community Support, and Alpha Testing.  This way we can discuss the problems, and also help with "End User" support a little bit.  This way the "End Users" are not coming to the Developer threads, and asking "How do I setup a server" types of questions.

 

We can work on Alpha Testing, work on documentation, and work on world build demos.  I think those 3 things will help the realXtend team the most.  These are very time consuming things, and just not something the core development team has time (or resources) to focus on.  So let the core developers work on getting the code written, and I'll just work on doing the compiles every 2-3 weeks (just someone send me a private E-mail every 2-3 weeks to remind me to do a new updated build).  I get busy sometimes, and as long as someone reminds me, I can do a new build every 2-3 weeks (build early, build often).  Then we can all upgrade at the same time (so that way we are all discussing bugs with the latest build.

 

So we are all on the "same sheet of music".  I'll get something done this weekend, and have something posted (with a new web page, and a new forum for Alpha Testers) by Monday morning.  Then I'll post a link here in the thread.

 

This way we can take our support problems over to a new area, and at least I can freely post the Alpha builds, without disrupting any of the realXtend core developers (and we can keep things off of the realXtend website, which Doeko just mentioned which does sound like a good idea).  I think we have enough "Alpha Testers" here at the moment, and you really have to have some EXPERIENCE or "be a pro" as Doeko stated, just to be messing around with Alpha testing.

 

So those that are "strong at heart" and willing to do the legwork, can head over to the Alpha Forums, but don't expect too much "hand holding" and the Alpha Forums will be more of "bug reports" and issues with the latest Alpha Builds.  This way we can give this feedback to the core development team, and we can also work on getting the realXtend WIKI updated from the fuscia color, over to the green color (tested/working).  We have a long way to go, before we get from Alpha into a stable Beta, and then get into a Final 1.0 Stable release (production).

 

At least the "Alpha Team" can help with testing, and help with working on "realXtend ModRex" Demo world builds.  Then if we can build maybe 3 or 4 really excellent worlds (regions) just demonstrating all the key technologies, at least we can get these posted back to the realXtend website (for download), and also have a working "Beneath the Sea" demo that will work with the ModRex Beta (when the Beta is finally released).

 

This way there will be a demo, the ModRex binary, and instructions/documentation for new users.  Plus at least we'll have a small Community of "Alpha Users" who can help lead some of the newcomers, and at least help/coach/teach them a bit.

 

This way the core development team can continue working on writing code, and they aren't bombarded with silly and crazy questions "newbie questions" like "How do I compile a binary?"  or "How do I build a server?"  or "How do I connect to the internet?"

 

These questions need to be taken off of the Developer site, and be moved over to a "Community" site (which the Alpha Team) can eventually help run. 

 

I guess I'll take the Bull by the horns, and but this on my list of "Things To Do" (my poor list keeps growing out of control), but it does seem to need to get done.  I believe it's the best way to move forward as a group/community.  This way we can keep the "newbie" questions off of the Developer's thread. 

 

Plus the Alpha Team can handle a lot of the "technical support" for the end users and early training.  Just till the Community gets large enough. 

 

I'll try to get something built over the weekend, and I'll post something back to the list with compiled binaries on Monday.  So Alpha Testers can sign up and can begin working on the Alpha Testing of ModRex and eventually the realXtend ModRex Community Support, and help with the realXtend "ModRex" demo builds. 

 

If enough of us work together, we can get the "Beneath the Sea" demo updated fairly quickly, and then we can begin working on maybe 3-4 demo worlds.  I'd like to see some form of "Naval Base" demo (with ships and working cranes, vessels, etc.) then also work on an "Airport" demo (with flying aircraft, ground support vehicles, etc.) and then also work on a "Railroad" demo (with a working rail yard, rolling stock, and even track switches, and working lights).  Then later we can work on advanced features like an Aircraft control tower, or Railyard control tower.

 

Everything will take time, but at least we can "demo" the advanced features of RealXtend ModRex.  I'll work on getting a small Community Forum setup for early "Alpha Testers" and get the ModRex binaries built.  I'll try to have something up by Monday.

 

The key to "Alpha Testing" will be to test the ModRex builds, and also work on a public demonstration of the realXtend capabilities of ModRex.  Once we have something good built, then we can just package it up nicely with a Beta build (with some good documenation) for the end users.  This is a big project, but at least we can help take some of the load off of the core realXtend team.

 

> Thanks, I hope you understand my point of view.

 

I completely understand your view.  I can see where you are coming from.  I believe maybe you, I and LC can work together on early Alpha Testing, and I'll setup a forum where people can download Alpha Team binary builds there.

 

At least we can begin working.  Leading development teams is like "herding cats".  Developers all head in their own opposite dierctions, and we get so busy that it's hard to worry about what everyone else is doing.  So let the core development focus on writing code, and anyone that considers themself a "Pro" can head over to the Alpha Team Community, and I guess I'll take on that project of working on Alpha Team testing.  I'll have something built by Monday, and I'll post a link and more information to the list by Monday.

 

Alpha Testing seems like the next "logical step" as we progress as a Community.

 

                            Thanks,

 

                              Mark

[Ryan McDougall]

On Tue, Feb 17, 2009 at 9:02 PM, doeko....@gmail.com
<doeko....@gmail.com> wrote:
>
> On 17 feb, 19:10, Ryan McDougall <sempu...@gmail.com> wrote:
>> On Tue, Feb 17, 2009 at 6:14 PM, Mark Malewski <mark.malew...@gmail.com> wrote:
>> I don't think there is any case of wanting "100%" done. I have been
>> working on open source for my entire professional career, and I
>> understand "release early, release often". However I also understand
>> that at this point reX is alpha software, and alpha software isn't
>> meant for users yet. It's meant for people with the stomach for
>> downloading from SVN, compiling it themselves, and helping with bug
>> reports. The source is freely available, and building it isn't very
>> difficult compared to any other project in an alpha status.
>
> I see where you are coming from, and I partially agree. However, there
> is an amount of users that, I think, has enough technical
> understanding to set things up and could contribute bug reports.
> However, they don't have the technical knowhow to do a full compile,
> or simply not the amount of time required to figure out how to do it
> due to lack of experience. I know I am one of those people.

Well if its only a matter of making a couple DLLs available, there is
no problem at all. However it was my impression that 1. more was being
asked for, 2. that once we hand out a DLL, it still won't work as
expected, and cause issues trying to support that DLL.

If I'm mistaken, its an easy fix to make.

Cheers,

[doeko....@gmail.com]

Well, I don't speak for anyone else, but what I'm asking for (and what
I think would be very much appreciated by many) is just that.

However, I must say, I actually have the DLLs and I am struggling a
little bit with how to make them work correctly, so if a very short
note on how to apply them to Opensim can be made available then at
least I am very happy. I also think that anyone with some sense will
not ask for support (apart from the below) regarding modrex as I am
pretty sure that anyone who even knows what modrex is, will know they
can not expect anything from it.

If a short note can't be made available, because it is complicated
(like, it takes more than 15 minutes for someone to write it), then I
guess I will just have to wait until something more final comes
along?

Also, let me take the opportunity to say that REX is the most awesome
thing to happen to virtual worlds (next to opensim of course) and I
think that the project will be very succesful in the long term. The
tools you offer are exactly what many have been waiting for, so you
can't blame them for being eager to start using them!

On 17 feb, 22:17, Ryan McDougall <sempu...@gmail.com> wrote:
> On Tue, Feb 17, 2009 at 9:02 PM, doeko.cass...@gmail.com

[Paul Fishwick]

Doeko
Have you seen:

http://opensimulator.org/wiki/ModRex ?

-p

--
Dr. Paul A. Fishwick E-Mail: fish...@cise.ufl.edu
Dept. of Computer & Info Phone & FAX: (352) 392-1414
Science and Engineering WWW: http://www.cise.ufl.edu/~fishwick
University of Florida (PGP Key available at above WWW address)
P. O. Box 116120
332 Bldg. CSE, Gainesville, FL 32611-6120

[doeko....@gmail.com]

Yes. To specify further my request for a setup note:
"Readable for someone who is not familiar with Visual C++"

Regardless, I'll have a look if I can make it work with the
instructions on there (as I do have it installed), but I doubt I'll
get very far!

> Dr. Paul A. Fishwick           E-Mail: fishw...@cise.ufl.edu

[Paul Fishwick]

It is Visual C# (which is a free download). For the Windows
instructions, let me know if there is anything that is confusing or could be
clarified. To my knowledge, the instructions are accurate, and one doesn't
need to know C# to execute these instructions. But, alas, yes, using modrex
does currently require building from source... They are trying to keep
up with
the opensim head, which recently is like trying to shoot at a moving target
(mainly due to refactoring..)

Dr. Paul A. Fishwick E-Mail: fish...@cise.ufl.edu

[Mark Malewski]

Ryan,

 

> However, if you intend to contribute your own resources to debugging
> "Under the Sea", and generate a list of regressions for the community,
> then we would eventually know where they all are, and we would be one
> step closer to having them fixed.

 

Yes, that's exactly what I'm saying.  Yes, we'll "contribute" our own time and resources to debugging "Beneath the Sea" demo for ModRex.  

 

I'll put out a "call" to all ReX Community members, and setup a Grid Server specifically for ModRex, and hopefully we can slowly begin working on developing some "demo" regions for ModRex.

 

I'd like to create maybe 4 demo regions.  The first region, being a "ModRex" Under the Sea update.  I hate to see Under The Sea get wasted, since RealXtend did such a great job on it.  So I would like to do an update for it, so that we can "give back" to the ReX Community.  This way the updated/upgraded demo worlds can eventually be downloaded (via the RealXtend website) once the demo worlds are finished.

 

I want to eventually create 5 working ModRex demo worlds that will "knock the socks" off of people, just to demonstrate the capabilities of RealXtend (ModRex). 

 

1) Beneath the Sea (ModRex update)

2) Naval Port Demo (with ships/vessels, cranes, etc.)

3) Airport Demo (with aircraft, ground support vehicles, etc.)

4) Railroad Demo (with Railyard, trains, railcars, switches, etc.)

5) City Demo (with skyscrapers, an advanced Metro subway system, taxi cabs, public buses, cars, trucks, semi-trucks, etc.)

 

It would need to be done in stages, but I believe we could create some very advanced demos, to show the advanced capabilities of ModRex.  I believe the "first step" would be to work on a "Beneath the Sea" ModRex update.  The work we do on this, can help as a "ModRex Community University".

 

So ModRex users can learn more about ModRex, and we can slowly teach others how to build, script, and use ModRex.

 

>Bottom line, this is a community of open source users and developers.
>It only works when everyone is lending their hand. Otherwise, I am
>afraid you'll have to wait for our hands to become freer.

 

I agree.  But I don't believe the individuals here just want to "sit on their hands" and wait.  I really do believe that everyone in the community here are just waiting anxiously, so we can begin learning, and begin documenting, and begin building (and testing).

 

We need to get documentation updated, need to get some building/testing done, and the only way to bring this project out of a "Fuscia" color into a "Green" (tested) color, is that we need Alpha Testers, and an Alpha World (where we can build, demo, test, and collaborate together as a Community).

 

I have servers, and I'll work on getting something built, and something up this weekend.  I'll have something for everyone on Monday, and I'll do an open call to anyone that wants to help Alpha Test and Alpha Build.

 

We'll need to create good documenation, and then we'll just use it as a "portal" to give feedback back to the development team.  This way we're taking all the mess out of the developer's message lists, and forums, and at least we'll have a "sandbox" for new users, and for "early adopter" Alpha Testers, so we can come together and learn/test ModRex.

 

> Lastly, if the only issue here is getting a compiled binary in your
> hands, I am sure Mikko can find his latest DLL somewhere. However,
> don't expect it to be smooth sailing. That'd be like expecting an egg
> to come out of the fridge boiled.

Ryan, yes a compiled binary would help quite a bit.  If Mikko can conjur one up, then please send it to me, and I will post it within an hour or two of you sending it to me.  This way the "Alpha Team" can begin testing, and begin documenting and testing.

 

I understand that we are still in early stages of development, and we are just asking for "something is better than nothing".  As we work on updating "Beneath The Sea" we will find out more about what is broken in ModRex (and we can post these bug reports, and also post "Feature Requests")  This way the developers are getting some feedback, and also ModRex is at least being thoroughly tested.

 

I'm going to setup a "ModRex" grid, specifically for testing the new ModRex.  I have some server equipment laying around, and I'll just try to get something up and running by Monday.  I'll try to setup a forum, and various other things... just so we can start to answer "newbie" questions, and setup a "ReX Community University" where we can teach newcomers how to build, how to use ModRex, and various other things.

 

For now, the Alpha Team will just focus on learning, building, and documenting.  Later, once ModRex reaches Beta, then the Alpha Team can begin teaching others.  (At least this way we'll have a small group of "experts" that can help with documentation, tutorials, and teaching).

 

I'll just put a call out to all "early adopters" to please come forth and join the ReX Community "Alpha Team".

 

This will take the load off of the developers, and the Alpha Team can eventually provide "Technical Support" to the new users.  This way new users are not bothering or pestering the core developers and the developers can focus their efforts on writing code, and not answering silly questions.

 

The "Alpha Team" can act as a bridge/gateway between the Community and the Developers.  This way the developers can focus on writing code, and getting ModRex finished, and getting the RexViewer started (and done).  The ReX Community (Alpha Team) can at least start the testing, documenting, and tutorials, and get the Alpha Worlds (demo worlds) built.  Just to publically demonstrate the advanced capabilities of ModRex.

 

Then later we can have the demo regions available for download (and package up an upgraded version of "Beneath the Sea" for ModRex).  This way new users will have something to look at and learn from.

 

> Also, let me take the opportunity to say that REX is the most awesome
> thing to happen to virtual worlds (next to opensim of course) and I
> think that the project will be very succesful in the long term.

 

I second the notion.  I believe ReX has the potential to become the best thing to happen to Virtual Worlds, and it's just going to take some hard work (and time) to slowly mature the product and slowly develop a large community of users that are familiar with ReX.

 

Those here, are just some of the "early adopters" and I understand the "growing pains" of getting a small (and YOUNG) project off the ground.  But I believe in time, that ReX could be extremely successful in the long term. 

 

I'll work on getting something a bit more "concrete" where Alpha Team members can congregate, and collaborate.  Plus a small area where we can post the ModRex DLL's, announcements, and documentation updates and building tutorials (and "how to" videos).  I'll try to have something up by Monday. 

 

I believe if we come together, and work together that we can help (as a Community) instead of just sitting here waiting on the Developers to do all the work (or finish the coding).  At least this way we are doing something productive for the ReX Community, and at least we'll have something for others to follow, use, and to demonstrate the capabilities of ReX.

 

                  Mark

 

P.S. If anyone is good in graphics design, please contact me and send me an E-mail.  I'm an engineer, and I have a hard time matching my socks let alone designing fancy graphics.  I may need a bit of help with some graphics and logos.  (Ryan, I'll post what is submitted in the Forums, and then the Community can cast their votes on a Logo, and possibly a "mascot" or "icon").  Then after the Community Votes, then I'll leave it up to realXtend "staff" to give the final blessing.

 

I registered the domain "rex-community.org" but please wait till Monday for anything to appear.  Nothing is "magic" and everything requires time, and hard work to get this stuff all off the ground.  If anyone here is in Graphics Design, then please contact me.  I may need a little bit of help just to get something off the ground before Monday (I always need a bit of help with graphics).  Thanks!

[Mikko Pallari]

Hi,

As Ryan said, ModRex is not yet ready for wider testing. It is still very Alpha.

To clarify the state of ModRex I quote Adam Frisbys blog (http://www.adamfrisby.com/blog/2009/01/modrex-02a-now-in-svn/): "Please remember, this is alpha software. Not like OpenSim 'it sort of kinda works' alpha, but really early code. It may break. It may trash your sim. It may kill small children."

Although this was from about a month ago and we have made progress since that, I still think that this applies.

One of the biggest missing features is multi region support. This means: no hyper grid, no grid mode, not more than one region in sim etc.

______________________________
Mikko Pallari
Server Technical Lead in RealXtend

ADMINO technologies
www.adminotech.com
www.realxtend.org

-----Original Message-----
From: real...@googlegroups.com [mailto:real...@googlegroups.com] On Behalf Of doeko....@gmail.com
Sent: 17. helmikuuta 2009 21:03
To: realXtend
Subject: Re: RealXtend Server 0.62 - Was: Re: [REX] Re: What is the status of ModRex?

[Lc]

We all know the actual limits of modrex.
What do we want is a build we can start working on :

1. creating contents
2. validating toolchains
3. make good documentations
   
   

The goal is not testing all the bells and wistles but starting working seriously on that product and I won't spend a week to understand why monodevelop crash as soon as i load osim+modrex


Sm

[Ryan McDougall]

On Wed, Feb 18, 2009 at 9:51 AM, Lc <lcc...@gmail.com> wrote:
> We all know the actual limits of modrex.
> What do we want is a build we can start working on :
>
> creating contents
> validating toolchains
> make good documentations
>
> The goal is not testing all the bells and wistles but starting working
> seriously on that product and I won't spend a week to understand why
> monodevelop crash as soon as i load osim+modrex

I think the problem is: if you can't debug why monodevelop crashes,
how will you debug it when ModreX crashes?

Cheers,

[Mikko Pallari]

I made a release of the ModRex to the OpenSim forge. See: http://forge.opensimulator.org/gf/project/modrex/frs/

And as mentioned in release notes: This is prealpha release. It may break. It may trash your sim. This release shouldn't be used for production.

[Lc]

this thread lead really to nowhere.
The problem is not the monodevelop issue but working together to create contents and docs.
What's the problem with that Ryan ?

[Paul Fishwick]

The sum total of all knowledge of how to make modrex is documented
in: http://opensimulator.org/wiki/ModRex

However, I noted in this release that there is the following folder: Physics
containing ModularRexOdePlugin.dll and RexDotMeshLoader.dll. There
was no communication or readme file indicating that these needed to be
built. Should I update the procedure on the above web page? I am not
sure what these modules do - when I add them to the Physics folder
under opensim\bin\physics, I don't see any changes. Is a new setting in
opensim.ini required?

-p

--
Paul Fishwick, PhD
Professor and Director, Digital Arts and Sciences Programs
University of Florida
Computer & Information Science and Eng. Dept.
Bldg. CSE, Room 301
P.O. Box 116120
Gainesville, FL 32611
Email: fish...@cise.ufl.edu
Phone: (352) 392-1414
Fax: (352) 392-1220
Web: http://www.cise.ufl.edu/~fishwick

[Mikko Pallari]

Hi,

The readme file is indeed a bit outdated. The purpose of this physics plugin is to add collision mesh support, raycast and volume collisions. Tuco has been mostly working on physics stuff, so he can inform you about the state of raycast and volume collisions. Collision meshes should however work.

[Tuco]

Situation report on the modrex physics:
- physics=RexOpenDynamicsEngine can be used in opensim.ini to use rex
ode physics
- mesh collisions should work. Uploading .mesh files also works.
- Not working: raycast is under work, should finish soon
- Not working: volumecollisions, opensim has some functionality which
is similar to this functionality.
- Not working: Disabling flying & sitting
- Partially working: max flight height.


- Tuco / LudoCraft

[Paul Fishwick]

Let's take this in two parts: 1) the fix to the documentation for those
brave souls wanting
to try out the latest, and 2) what "collision meshes" mean in terms of
the user experience:

PART 1:

It looks like I need to add the following. Please correct if you see
mistakes:

Instead of 3 preexisting projects needing to be added to the OpenSim
solution, now
there are 5. The two others are: RexDotMeshLoader and
ModularRex->RexOdePlugin.
Then, once opensim is built, these two binaries need to be manually
located under
c:\opensimcheckout\bin\physics.

Is this correct? And are there new missing instructions such as new
settings in
OpenSim.ini ?

PART 2:

The feet of my avatar are still 6 inches above the ground. And when I
take my
OGRE fish and make the Collision Mesh the same mesh "fish", my avatar does
not collide with it. Collision of the avatar with prims has always
worked (at least
for the past 2 or 3 weeks). So, what functionality has been added?

Thanks.

-p

[Paul Fishwick]

Tuco wrote:
> Situation report on the modrex physics:
> - physics=RexOpenDynamicsEngine can be used in opensim.ini to use rex
> ode physics
>

aha

> - mesh collisions should work. Uploading .mesh files also works.
>

uploading .mesh files has worked for a long time. Let me now try the
inter-mesh collision.

> - Not working: raycast is under work, should finish soon
>

ok - tell me how one can use raycast. I understand the principles since
it can be used
in 3D selection and ray tracing, but I would like to know how a user can
employ it.

> - Not working: volumecollisions, opensim has some functionality which
> is similar to this functionality.
> - Not working: Disabling flying & sitting
> - Partially working: max flight height.
>

ok, when these work, please post to the group.

thanks

-p
>
> - Tuco / LudoCraft

[Paul Fishwick]

I am now using RexOpenDynamicsEngine, although I confess that I don't
see how this is different than using OpenDynamicsEngine since the latter
seemed to work fine when creating physical objects and throwing them
around over the terrain. Note: I am using Meshmerizer for meshing instead
of zeromesher. Not sure of the difference, but most people seem to use
Meshmerizer.

Observations: the avatar still walks 6 inches (or so) above the ground, and
when I set the collision mesh of the fish to the same mesh (fish),
collision of
my avatar with the fish does not work. Should it?

-p

[Ryan McDougall]

I am sorry, I have completely failed to understand what the problem is...

Cheers,

[Mark Malewski]

Mikko/Tuco/Ryan,

 

It seems that you are hard at work, and I can see the progress that you are making, but just as you stated previously... lots of work still needs to be done (and ModRex still has a long way to go before it's ready for Beta testing).

 

I believe many are confusing Alpha Testing (Debuggers) with Beta Testing (mainstream content developer Testers and Users).

 

I see you've been moving quickly, but it's hard to even tell when we would even be close to a Beta. Thank-you for posting the ModRex binaries to the OpenSim forge.  It will at least give a few early adopters something to cry about.  ;-)

 

> And as mentioned in release notes: This is prealpha release. It may break. It may trash

> your sim. This release shouldn't be used for production.

Everyone has been properly warned.  Let the herds do what they will with it. 

 

> We all know the actual limits of modrex.
> What do we want is a build we can start working on :
>
> creating contents
> validating toolchains
> make good documentations

I understand what "we ALL want", but after reading Mikko's previous comments, it's pretty safe to say that... ModRex just isn't ready yet.  The warnings and disclaimers should be enough of a warning to fend off the weak at heart.

 

The strong at heart will tinker, and I'll setup a separate site for "gripes and complaints" so that we are not bombarding this whole thread (with Alpha Testing complaints).  Personally, I will probably just start with one single region, and try to see if it's possible to move "Beneath The Sea" from the 0.4 Server over to a ModRex build, and then from there do what we can to slowly implement the ModRex features (to try and get "Beneath The Sea" to work properly under ModRex).

 

The developers still need time to work though, and as we upgrade the "Beneath the Sea" to ModRex at least it will give us an opportunity to do a little bit of testing, and give the developers a bit of feedback. 

 

I may give it another few weeks, and let Mikko and his crew continue their work.  I know everyone is in a rush to get something "usable" so we can begin creating content, begin demo work, and begin good documentation (and tutorials / video tutorials), but it seems we may have to just be patient and just give the core ModRex developers more time.

 

I can see the progress from the ModRex WIKI pages, but we still have a long way to go on updating documentation and most users don't realize how "unfinished" and "unstable" ModRex still is at this point.  So the "PRE-ALPHA" binaries have been posted (as Mikko stated) to SourceForge.  If you are bold, and strong at heart... give it a try, but just don't cry or complain.  (You've been fairly warned).

 

At this point, all we can do is give Mikko and the developers more time.  They are hard at work, and we can see the progress they are making, but they really truely need more time to slowly mature the product (so we can begin truely testing ModRex). 

 

> I think the problem is: if you can't debug why monodevelop crashes,
> how will you debug it when ModreX crashes?

Ryan has made a very good and very valid point.  As Mikko has stated previously, this is almost not even an Alpha, this is a PRE-ALPHA.  ModRex really does need more time, and more hard work.  

 

I understand that everyone is anxious, but messing around with a PRE-ALPHA is almost like beating your head against a wall.  Code still needs to be written, and tons of documentation needs to be updated.

 

The developers need more time, and I don't think everyone "fully understood" the infancy of the current status of ModRex.  The progress charts (with implemented features) seemed a tad bit misleading to some (at first glance) probably because many are anxious for a "usable build" and I was personally shocked at how far ModRex has come and how quickly the developers are working. 

 

A ModRex build has been posted, if you can't wait any longer then dive in and enjoy the heartache, and I'll setup a Rex Community Area where you can share your comments and thoughts (and struggles) as you tinker with the PRE-ALPHA release.

 

As we move closer to an actual Alpha, then we can begin truely working on content creation (as ModRex becomes more stable, and more mature).  I can see that a LOT has been done so far, but as Mikko has stated previously... "It may break. It may trash your sim. This release shouldn't be used for production."

 

Paul Fishwick stated:

 

> The sum total of all knowledge of how to make modrex is documented
> in: http://opensimulator.org/wiki/ModRex

> However, I noted in this release that there is the following folder: Physics
> containing ModularRexOdePlugin.dll and RexDotMeshLoader.dll. There
> was no communication or readme file indicating that these needed to be
> built. Should I update the procedure on the above web page?

Paul, yes... please do!  The OpenSim ModRex WIKI page does need some work.  ModRex Documentation needs LOTS of work.  Tutorials (with step-by-step procedures) still need to be written and the ReadMe file needs some updating as well.

 

I wanted to keep a vanilla OpenSim Grid running (for content development), but also wanted to connect a separate ModRex region to the Grid (for testing and "sandbox" purposes).  Would this be possible?

 

>> One of the biggest missing features is multi region support. This means:
>> no hyper grid, no grid mode, not more than one region in sim etc.

Mikko, I understand ModRex is still under development and this may sound like an incredibly stupid question, but is there any possibility of running both ModRex and non-ModRex regions on the same Grid?  Has any thought gone into "backwards compatibility"?

 

Could something like this be done?  Is this a feature that could be added?  Is there any possibility that ModRex could be implemented on the same main Grid Server and that ModRex could be enabled or disabled on a "per region server" basis (in the OpenSim.ini file?) while still allowing "non-ModRex" region servers to connect to the main ModRex Grid and maintaining backwards compatiblity with both ModRex and non-ModRex region servers on the same Grid?

[Mark Malewski]

Paul,

 

I believe good well written tutorials, and good "step-by-step" procedures and good documentation will be key to implementing ModRex and widespread ModRex use.

Please document everything as you go, and please document your procedures and thank-you for taking the time and helping with the WIKI pages.

 

> ok, when these work, please post to the group.

Tuco, please do post back to the group when these features are working, just so we can begin testing them and documenting them, and begin updating the WIKI pages.

 

> ok - tell me how one can use raycast. I understand the principles since
>it can be used
>in 3D selection and ray tracing, but I would like to know how a user can
>employ it.

From what I understand, Ray-Casting has a very "blocky" look (limited by one or more geometric constraints) and is used for very "flat" or "square" or very simple geometric shapes such as a flat wall or floor tile (Ray-Casting would be good for walls, floor tiles, etc.), and Ray-Tracing is generally used for more "odd-shaped" or "any-shape" renders (round or odd shaped objects). 

 

This is a brief tutorial on raycasting:

http://www.permadi.com/tutorial/raycast/

 

I understand the concept of raycasting as well, but yes we would need some good documentation on how to employ the raycasting.  I understand raycasting (in theory) is much faster than Ray-Tracing, because rendered images are not stored on disk, but normally only the map is stored and corresponding images are generated "on the fly".  (Wolfenstein 3D, Shadow Caster, Doom, etc.)

 

I supposed this would be good for generating flat objects, walls, floors, ceilings and cubes (texture mapping walls or "floor casting" or "ceiling casting"). 

 

But from what I understand, I believe Tuco was referring to something more along the lines of a "Physics.Raycast", which would be used for "hits" or when a ray intersects with any collider.  I supposed it could be used to cast a ray against all colliders in the scene and return detailed information on what was hit.

 

Something along the lines of this:

 

http://unity3d.com/support/documentation/ScriptReference/Physics.Raycast.html

 

I really don't design games (so I apologize for my ignorance) but from what I understand, I supposed a Physics.Raycast like this could be used in gaming to show information on hits and could also be used to cast "damage" displays on objects.  For example if you punched a garbage can, that collider information could then be used to map and display a damaged garbage can, or display a black burn mark or a damaged overlay texture.  I supposed something like this could optimally be used to show damaged cars, damaged objects, or damaged vehicles.

 

           Thanks,

 

             Mark