Urgent Feature

Hi old friends, I bring news from the first life,



If you don't sell your sever software and let us connect to the grid, people are gonna steal your schtick. So compile up a server, make an installer, put it up for download and charge us $100 a year for the account we need to connect to the grid. PLEASE HURRY.

Any server running on your system would not be part of the main grid and thus you would not be able to transfer items back and forth.

The biggest advantage of SL is it's hosting. There are better software packages you can get for 3D worlds, but none of them have anything close to an economy or property since they are totally open and you can copy everything and make your own money.

There is something said to having a local server, but having a local mega cache server would be better.

Yeah, I honestly don't think Linden Lab is in any grave danger of someone stealing their schtick, any time soon anyway. It would be a monumental (and cost prohibitive) task to reinvent the community, economy, content, and the many other things that are unique to the Second Life grid.

Mitch Kapor and other executives at Linden Lab have made no attempt to hide the fact that ultimately the server will be open sourced - it's "on the road map" but not in the immediate future.

This sky, too, shall not fall.

zk

Any computer connected to the internet could technically be a part of their grid if they allowed it.



They currently have the monopoly on hosting, to me that is not an advantage.



Can you name any? I did a small amount of work on the Open Source Metaverse Project and I can tell you that the server part is sooo much easier than the client. When the client is open sourced, I promise a public domain server will be released within a year that lets people run their own grid with the Second Life opensource client. And TPTB can prevent this by releasing a closed source server that doesn't handle cash transactions or run scripts, and charging us a fee to connect our server(s) to the grid.

Seriously if you open source the server, it'll be reverse engineered in the public domain, google will host for free, all within 5 years. Mark, my, words.




Being "open" has nothing to do with being able to copy everything and make your own money. Being the admin on the server that holds that information is what gives you those rights.



Agreed, so if there was, oh I don't know, some kind of service available that would let business invest their marketing and hardware to the secondlife grid rather than build their own... Hmm... Like a "server" you could run and a fee you could pay...

But if the only thing available is some public domain source code to build my own, and it's all going to be user generated content anyways, and I'm frickin google, then nothings going to stop me.

Sorry but most of the code/smarts/work are in the Server side. The open metaverse server is nothing like an SL server.

It's like comparing 2 kinda similar but not really the same things and saying that because 1 is simple, the other must be also.

Could someone dev a server from working with the client source, yes.

Could they do it in a year, yes probably.

Could they do it in a year, and have reverse engineered all the little things, Unlikely.

Which means that it would kinda work, but not really, and without the ability to connect to the "Linden" Grid isn't going to get far quickly.

You need to replicate the Social Environs as well.

SL software while an amazing piece of work, is _NOT_ they key to why SL works.

Personally I would love a local to me server for content development purposes, of course I would want to be able to transfer the completed content into the grid easily as well.

Another thought, a SL server is not _A_ server, it's actually needs at least 3 maybe more for it to work, (sure you could combine them into one machine for a low use server/dev server)

My suspicion is that once they do get around to releasing the server, open source or otherwise, there will be a rather involved application process - and a laundry list of some pretty hefty minimum requirements for hardware, Internet connectivity, etc - before it is approved to connect to the grid. Not unlike the current process for connecting a new server to the EFNet IRC network, but even more stringent. It isn't likely that "Average Joe Citizen" is ever going to be able to run a grid-connected server.

It would also be woefully slow at connecting to asset servers and such, even if you could get an internet connection approaching the speeds of a semi-decent local network JUST to connect to the asset servers, then latency would still slow it down. Add on top of that you'd need to share that connection with all outgoing traffic to users, you'd either have one slow and nasty simulator, or a very costly one unless you can get yourself a server farm and bulk bandwidth.
If you think packet loss and such is bad now, imagine what it would be like on a machine external to LL's network where packet loss is occuring with central resource servers!

Regarding the amount of code, who knows how many lines are "code" are in a SL Server vs the SL client. When I said that once a client was available it will be easy to write a server, that was not a statement about the amount of anything in the SL server vs client.

A server is not displacing anything unless there's a robotic arm attached to it (there probably is a server at LL with a robotic arm attached to it but that's beyond the scope of this thread). So IN FACT, both servers are doing 0 Work if you measure work in Joules. However, I once connected a home automation device that used a old latching relay (moving parts) to an object in Second Life. So technically their server did some work... Cool. But I didn't use it for very long so all the work I did in pressing keys to write the code (which happened on the client side) amounted to more than that of the server.



Not really, It's like saying I find client side programming much more difficult than server side. There is more more work involved, more energy expended, more knowledge required. If you don't agree with my opinion based on my experience, why don't you state your opinion based on your experience?



SL is USER GENERATED CONTENT. Nobody has to replicate that. You draw in users and they build the content - care to compare how quickly the user base grows for oh say a Microsoft game compared to how quickly the user base for a LL game grows? The speed your user base grows has little to do with your code and much to do with your wallet.



Just for the sake of semantics, you could combine them into one executable and make them run on the same port. But I wouldn't... There's a lot more than 3 servers, the information is published here



I totally agree. Closed source server, the people running their own servers subscribing for a grid connection to LL, will not be your average Joe. Maybe EA will buy 10 islands and start a continent. Maybe a year later the continent will be 1000 sims bigger. Maybe a year after that, 10000 sims bigger. Alas, we'll never know...

For an open source server, the people running their own grids will not be your average Joe's either. Let's say EA realizes how cheap it would be to run a grid and charge users $x a month to get a playable account and let users run the servers and charge them $100 a month to connect the servers to the grid...

I once believed SL had the potential to become the defacto gaming development environment. That eventually, Quake 7 would run on Second Life. Nobody really agreed and I think I know why. That was a VAST understatement. Today I believe that SL can one day be to media what a web browser is to knowledge.

History will read: "Phonetic Alphabet, Priniting Press, Telegraph, Second Life".

I understand that features have the potential to arrive more quickly with when things are open sourced. But there is no real guarantee of this. Just like there is no real guarantee that my fear of unstoppable Second Life clones will surface. I just worry about it is all. You can do what you want with this information.

I'm not the only one that wants to sell SL server hosting. There are lots of people ready to sign up for this but WHAT THE BLEEP DO WE KNOW. There's no information on the roadmap about it, so all I can assume is you are going to actually ENCOURAGE people to make products that compete with SL as long as they share their source code.

Anyways, wouldn't more time be spent on contributing to your open source initiative if a diverse group of businesses were able to add features their customers wanted without waiting for you?

So I thank everyone in this thread for playing devils advocate and offering their opinions based on your feedback, I would like to restate my original post.

Dear LL, "I will patiently await the open source server, but how soon after the server is released will be be able to run our own servers that connect to the SL grid?"



Good the next step is to attempt to falsify that theory. Setup a SIM that runs in, ohhhh, say.... Ottawa, Canada and let people visit it to see what kind of performance they will see when/if the service is ever provided.

While the client has gone open source, I wouldn't be surprised if any server side software that is released will come with a hefty price tag.