It's worth noting that what we do in practice isn't solely dependent on the open-source code. There are database and network constraints as well. Yes in one area may well be no, because in another.

Towards "one shard fits all", I was wondering about inter-shard communication on different networks. It seems as though it's possible.

W2003 server as network router

Questions of ownership / cost / uptime would come later.

Behh... the fact is you might not want to add such code if it looks like the whole start-up process would have to be rewritten to implement seamless in-game server disconnections/connections.

Looks to me like the network currently connects to a server at a very very early stage in startup. If you can easily interrupt the connection and change the address I don't know. Plasma I don't think has ever had in-game server switching... this could make it even harder.

I'm not trying to kill everyone's hopes here. But when people are quick to jump to conclusions it usually ends badly for all parties involved.

But anything is possible with enough time, coders and lack of sleep.

_________________
Guild of Writers Councilor
PyPRP2 Developer
Plasma Hacker

IRT OP:

Trick question, it doesn't matter as long as a given client (or server!) runs the game "legally" (i.e. no cheating or security vulnerabilities).

I don't claim to be an expert on these things, but I've been following another game that went through much the same troubles that Uru has gone through. Neptin, you play Descent 3, you've probably heard of it. It's called Freespace 2. It was a space combat sim released in 1999, and it is considered by many to be the best space combat sim ever made (not necessarily the most realistic, but in terms of fun factor, there's nothing better). Unfortunately, lack of marketing meant that not a lot of people bought it. The developer, Volition Inc. now works for THQ, making games like Saint's Row. The publisher, Interplay, makers of the Fallout series, went bankrupt soon after, and are now just starting to get back on their feet.

But anyway, the source code to Freespace 2 was released in 2002. Very soon after, a pair of programmers who were fans of the game thought that it would be better for any improvements on the engine to be made by a single group, rather than having a bunch of independent versions of the game that were incompatible with each other. So they started the Freespace 2 Source Code Project as an umbrella group to work on the code. They've updated the engine again and again (I believe they're at version 3.6.10 of the open-source engine) and built up a large community of programmers, modelers, artists, and level creators. In its open-source form, the game, originally released in 1999, could hold up against stuff on the Xbox these days. The graphics and such have been updated while keeping the core gameplay intact. And they've made the game engine modular, so that you can toggle the new features on and off. And many other groups have used the FSOpen game engine to make their own freeware games. You can find out more at their website: http://www.hard-light.net/

So, my question is, why not have a single ongoing project like the Freespace 2 community? A sort of ongoing development process? Everyone around here seems so paranoid and hostile towards everyone else when it comes to coding changes. Multiple game servers with different content are fine, but the idea of multiple engines is a prospect that I find pointless and inefficient. You can make it modular, so people can turn new stuff on and off. People like the old KI? Fine, they can use it. People want a new KI interface? They can do that, too. (I think... I don't know much about programming, so don't flame me, calmly explain to me what I got wrong like you would to a small child. )

Seriously, people wondering how to go about the open-source updating: Check these guys out. It's worked for them for the past seven years, and they've done amazing things with their game.

We're all friends here, right? Let's work together rather than against each other.

(If I've completely missed the point, let me know. )

_________________
"There is no Frigate like a Book To take us Lands away"
--Emily Dickinson

Go show your support to the makers of the Myst film!
www.mystmovie.com

Quote:

So, my question is, why not have a single ongoing project like the Freespace 2 community? A sort of ongoing development process? Everyone around here seems so paranoid and hostile towards everyone else when it comes to coding changes. Multiple game servers with different content are fine, but the idea of multiple engines is a prospect that I find pointless and inefficient. You can make it modular, so people can turn new stuff on and off. People like the old KI? Fine, they can use it. People want a new KI interface? They can do that, too. (I think... I don't know much about programming, so don't flame me, calmly explain to me what I got wrong like you would to a small child. Laughing)

It's mostly because it's going to happen the way you mention.

There really are other people like me quietly skulking around that agree with the general feel of what your saying but don't post it because all the planning in the world right now is just forum fodder. The open Uru site has some good rough draft ideas, and the guild sites have the odd thread or two devoted to getting MO:UL running again. Higher resolution base artwork, a real KI upgrade, modular design updates as needed, ect, ect, ect..... A lot of people agree with you Dr. Worm it's just people that can do something or feel they can anyway are working off the forums not on while they try to prepare.

It's a bit like trying to see the shape of something when all you can do is touch it in a dark room. Rather then try to figure out how the programming of Uru works; Why people are not holding coding classes I don't know, people are learning as many Uru'ish things elsewhere as possible. All in the hope that when we get the sources that the concepts if not the actual coding will transfer over intact.

_________________
Waymet

Papa Smurf wrote:

[…] Now we can control what runs on a shard so it would be up to the shard owner(s) to decide what, if any, KI mods will be allowed. Since this is the case, and since I know that the AdminKI was not only a good tool but a source of endless (and mindless) fun when bored, I would allow all users to have one. […]
[…] I know there would be a lot of people who would like some controls over their own Ages.

One can control what runs until someone modifies a version of the ‘client’. The direction in online games is to move to a standard 3D viewer that is hoped to become as ubiquitous as the Internet browser. SL has over a dozen special purpose clients. For WoW there are client plugin’s and hacks for various special uses. The same is true for other games too. Someone just developed a plug in for Internet browsers that will run WoW and other games in the browser. I think, we are likely to see variation in Uru clients. It was there, in some measure, with UU. I expect it to be the same in MOOS.

There are some great pictures of the AdminKI in action on UO. What people will do when bored… That UU got boring and lonely is a concern for me. People don’t do boring and lonely, at least not for long. They can play the stand alone games and get the lonely. Boring and lonely induces people to move on and that would mean an ever shrinking player base.

Chogon: … UruLive can still be great place to live.

I would love to have lived in a bevin. There were all those doors one could not open. In many games people have homes. Oblivion, SL, THERE… In a number of others one does not. I like having an in-game home. My Oblivion home is loaded with loot. I’m not sure how much that may affect game success. I do know it (renting property) is becoming a popular means to get financial support for various RPG’s. Serious players rent, transients trying the game out don’t have to.

Christian Walther ‘s shard definition: Definition: Uru Shard
Clear and eloquent definition. (I linked for the search engines).

MustardJeep wrote:

[…]It's a bit like trying to see the shape of something when all you can do is touch it in a dark room. Rather then try to figure out how the programming of Uru works; Why people are not holding coding classes I don't know, people are learning as many Uru'ish things elsewhere as possible. All in the hope that when we get the sources that the concepts if not the actual coding will transfer over intact.

Writing ages is not a trivial skill. I see the learning curve as very high. For an Uru age I also see it as very steep. There is so much one must learn to get even a simple age running. Look at the GoW wiki and threads in the forum. It’s a book.

This complexity is even truer for the server and client engines. I expect a few various trees in game engine code development but not dozens or hundreds. I also think there is more collaboration going one than we generally see discussed on the forums, a friends with similar interests thing. I doubt ‘how many Uru’s’will reach problem proportions because of the complexity.

While I do not see a way to reduce the height of the learning curve for ages I do believe it is possible to flatten the incline and have intermediate results and not have to get an entire age working first.

Many of us can build 3D models, textures, and code object and AV animations. In many of the other games modeling and AV animation is greatly simplified flattening the learning incline. Learning to make a texture look good in a game is basically the same in all games. The design cycle in many games is reduced because most of what we have not yet started building for MOOS is in place and working, speeding up the learning process. So, we can do trial and error for a single aspect or item and see immediate results. Plus being able to show something to another in game and get help is very handy.

For example, Jandai called friends in SL to help her test a puzzle for difficulty this week (a thing I’ve seen in Harry Potter movies that she has made into a neat puzzle). In less than a half hour a half dozen people tried the puzzle and she had her answer. There are several dozen if not hundreds of Uru fans learning to build 3D models, make clothes, furniture, animate things, animate AV’s (I made my first karate kick animation, lame as it is), build game puzzles, create storylines and make them ready for players.

I am with MustardJeep on the learning thing. Why more people are not learning, I don’t understand. Many claim to love Uru. But the number that contributes to the survival of Uru is way less. I wrote a beginners tutorial on programming for the GoMa. In the third lesson there is a graphical interface tool that anyone with drag and drop skill can learn to use (it’s an MIT thing). The factions of Uru people in THERE and SL are learning and growing. It is easier to learn concepts in those environments plus there are live classes for all time zones.

One is either learning and planning to help or waiting… how many Uru's is not a problem.

_________________
Nalates - GoC - 418 - MOULagain: Nal KI#00 083 543, Nalates 111451 - Second Life: Nalates Urriah
Guild of Cartographers

Nalates! I like the idea of renting and paying. Let's open up a possibility to rent islands and buy bridges and other relto addons in Yeeshas little shop in the clouds. (yes I know that you think about real gameplay changes that may advance the URU concept, still...)
Yeesha aka Cyan Worlds could let you rent a few acres in the sky beside your relto. (for a few bucks Yeesha also sell you the bridge page to it) And in the little shop in the clouds one could buy an extended bridge dock thingy to have dance partys on (with one of those disco light systems that is synced to a teamspeak channel). Everything is sold as pages, and you can also trade pages with friends.
– I need a crystal ball for my party over the weekend.

Hey I'm serious!
Only cyan may touch the relto maybe, so that is Cyans cash, but if open source agreement permits third party server owners to make money selling gadgets let's see them open the "UndergroundMart" and peddle alternatives to relto.
I don't see Cyan actually doing it, but that is their loss.

_________________
--- in our hearts our dreams are still the same ---

I'm just waiting for all the full class information... I want everything to be available at my disposal.

_________________
Can you withstand the gaze of the Eye of Eternity?

As far as this discussion I believe one should do a case study on BZFlag. It is the most portable game in terms of clients and yet all are interconnected with one portable code. Link: http://bzflag.org/

The program runs on Mac, Linux (Ubuntu,Sabayon,etc.etc.), and Windows. They use a central server to authenticate users but the servers themselves are all user provided.

_________________
I'm a URU Complete Chronicles Owner!

Zero, one or many databases?

I think there'll probably be a few different shards out there, and that some will be better, and more popular, than others. It's what's likely to happen in practice - I'm expecting it to happen, even though many of us would prefer to have everything integrated in one place.

There are some good and bad sides to having multiple shards.

The good is that:
-We have backup options should one shard fail, and can adapt more easily in that sense
-The multiple-shard system promotes competition among shard owners to provide better service to their members and become dominant shards.
-The multi-shard system gives players multiple choices, ranging from Cyan-style, tightly managed, consistently high-quality Uru to largely uncontrolled, Second Life-ish Uru which allows submission of flat gray planes as worlds.

The bad is that:
-Things may be split up in such a way that each shard has different stuff and different people in it. This can lead to the painful choice of which shard to join, if you only can afford to join one (non-free) shard.
-It may be difficult for players in one shard to interact with players in a different shard.

It'd be nice (as others have said) to find ways to integrate the many shards together so they feel like one shard even if they're on a bunch of servers owned by different people.

I may have ruffled some feathers by suggesting there may be development of content that is shard-exclusive (This age belongs to OUR shard and you can't have it!) but it's a possible thing that could happen.

I'd actually prefer that not to happen, but I still think the basic idea I was submitting - Cyan setting up basic production directives, story elements, concept art, puzzle designs, etc, and having the fans flesh them out into full ages, is doable. The trick is to figure out how to get Cyan to go along with that. It means, almost by definition, that there'd need to be a way for us to pay Cyan for such content to be given to us to work with and develop.

That'd mean an official Cyan shard, or an alliance of shards (an idea I like) pooling resources to work with Cyan.

Here's a nifty thought - what if, even though we have many shards, they're all integrated and they pool content?

What if there were a dozen or more shards, and all of them were set up so that people from any one of those shards could visit the others? Suppose that a fan-run alliance of shards were to share content, and work with Cyan to further Cyan's story (the initial designs by Cyan, the rest of the development by fans), using funds pooled among the different shards?

This sort of scenario offers a good mix of all the above good qualities of a multiple-shard system without the downsides, I think.

It looks like multiple shards will happen, but maybe we can connect them and organize them well enough that we can avoid the pitfalls that we'd risk by having a multi-shard system.

_________________
Matthew L. Hornbostel, creator of Panoramic Worlds,
a Myst fansite , and
various other stuff.

Quote:

What if there were a dozen or more shards, and all of them were set up so that people from any one of those shards could visit the others?

If you want the data to be persistent across all shards, the likelihood is that you have one database (vault) and thus... one shard.

http://alcugs.almlys.org/AlcugsArchitecture

For a bit on how the services work in Uru, Ironcially most of the gear we need is already there, they just need some modification to maybe work in a multi homed way

Reading the vault section is intresting. Notification messages down several hierarchy trees, shouldn't be too hard to transport that across servers.

It's been a long time since I've even read anything on this forum, let alone posted. I stumbled across this while trying to find info about Shards. Because, much to my surprise, I've found a living Shard! I thought they were all shut down, and none had been resurrected. Well, anyway, I'm about to join a Shard for the first time ever, and here I am seeing people arguing about how best to switch between shards. Plural?! OMG!!! I'm so excited I can barely contain myself! I have no idea how best to solve the problem permanently, but here is what I'm going to do. I'm going to back up my UCC, with the stuff I want, etc., THEN I'm going to apply the patch for connecting to the one shard I know how to get on. Then I'll back up whatever changed to a completely different archive. If, for some reason, I decide to connect to some other shard, I'm planning on replacing the current shard stuff with that earlier shard-less offline backup, and patching for the other shard. Then backup that into a completely separate archive. I have no idea how well this plan will work, I still might wind up with a huge amount of data being duplicated with only very slight differences, I dunno. But while you are all trying to figure out some kind of multi-shard thing, I'm going to be doing this.

I did have an idea, however, just now, about using some kind of KI/Nexus patch to connect to a shard. Let the system be offline until you go to the Nexus, and select a shard to connect to, and then everything gets brought online with that shard.... I dunno, it's probably a stupid idea. Have fun hashing it out.

So far my plan is working great, the patch only needed 9 megs of space, the archive shrunk the changes down to about 2 megs. That's it. Now I'll see if it runs. Well, the patched client runs, and downloads updates, etc. I'll get back to you once I actually start PLAYING, sheesh.

.shorah gah bigtotee b'shehmtee