A new server is being developed: Kaya.gs

[Kaya.gs]

I dunno, I still like this project, in spite of having some reservations.
As many have said - another server cannot be bad to the community, unless the community lets it.
And the worst that can happen is that the project will fizzle out, which I hope it won't.

Still, what goes through my mind is - how will Kaya.go compete?
I mean... I don't mind supporting it because it is a good idea, but why should I play there? I bet I am not alone asking this.
Go servers are like bars. Nobody wants to sit in a bar alone, so people go where people are. But how to get the initial people in?

Much like anything social. That is a challenge every current server went through and many others have failed to do.
We have plans on how to get critical mass, which is enough to play. We expect that when we get enough players to make the place able to get a game, and we have more features, the users will flow more naturally.

The really different thing I have seen is the embedded (streaming?) video capability. But videos are easy these days, and if this is to be the differentiator, pretty soon each server will have streaming video. Plus - videos have an overhead - somebody needs to shoot them. And for that to work, the shooting needs to be consistent and dependable, which usually means money involved, contracts, sending people to events, etc. If nobody at the next Congress videotapes the top games, this capability is
meaningless.

Actually i agree. Between monday and today i got several different video embeds ,and in our develompent server you can chat and view videos. You should ask all the other servers why they didnt do this, when the technology to make it easy was available for what..almost a decade?

What else? A kick-off tournament with good prizes can only get you so far before people go back to where they were before, and you need cash to do that. Luring in some friendly pros would be the best ticket, but I am not sure I have heard any plans for that. But this would certainly work... although it would also cost, probably.

Big flashy tournaments like Wbaduk is not our style , and its not long term fruitful. Now, user organized tournaments, thats another thing. People will make tournaments for themselves, their clubs, their associations. We will add a feature to schedule games, so you can agree with an opponent on a time, and if you both dont show you u lose.

With only 3 of the 20 features we have thought of for tournaments, we will already provide the best Online Go tournament experience out there.

The other good thing about Kaya.go I hear is that it will be web-based, rather than forcing the user to download a dedicated client. I like this idea, sort of like Yahoo!Go with better rating system and nicer interface. Is that enough?

The accessibility is a very clear and powerful feature. I remember signing up in Kgs instead of IGS, because i didnt understand the latter's registration form. Having a web-client for us is not so much for that accesibility though, as much as the other powerful web-related tools we will get.
Figure that part of the application will be open source, and the architecture we have will make it super-flexible so people can develop their own features.
The posibilities are limit-less.

When I look at the successful servers, there seems to be a pattern of filling in a void, a need. NNGS filled the need to people unhappy with IGS. KGS filled the void after NNGS folded. Other servers filled their respective niches and so they succeeded. But usually they already had a player base waiting before they started.

Then there are the other servers - all the less-than-successful ones. They usually tried to break off a chunk of somebody else's pie, compete within a market that is already pretty satisfied. This is what I see Kaya.go will be doing. And it is not clear to me why people should go play there rather than stay on KGS/IGS/wherever.

You are having this doubt because you are not using Kaya.gs. The things you will have available are just nowhere else. Its not a matter of comparing, its a change of paradigm for Go servers.
THats like comparing Facebook to Myspace, when there was no facebook page. You would have thought the same, what can FB give me that the already established Myspace community doesnt have?. A relationship status flag?



Your concerns are natural Bantari. Considering the ship has sailed long ago, id think of new and awesome features and suggest them so they can be considered as early as possible for Kaya. So just think of the wildest things you could use in a server, and put them in the feedback section .

Regards!

[jts]

I wrote three suggestions but it timed out and ate my post. Hopefully I can condense my thoughts and still be helpful.

1. The name/acronym choice is unfortunate. IGS, OGS, DGS, NNGS, KGS, and ... KGS? Even you're speaking, Kaya G.S. still sounds like K.G.S. So why not keep the theme and logo, but use a different name for Kaya?

Torreya Go Server
Hi Go Server
Fei Go Server
Picha (?) Go Server
Yew Go Server (Marketing slogan: "Why YGS?" Or: "What can Yew do for you?")

2. As clunky as Java can be, KGS is an epitome of spartan, effective design. It makes me pine for the simple, lightning fast applications of yesteryear. Please don't make a go server that is some Frankenstein combination of MySpace, YouTube, and Yahoo. Think Bauhaus, not Baroque.

3. There was discussion in another thread about the rating system. Why not throw ranks out the window? Use an algorithm like KGS's to predict a fair handicap for the first time two players meet, and then after that adjust the handicap like the old-school Japanese tournaments, considering only that pair of players. Different social clubs within the server can hold tournaments and award some honorary ranks to the winners, if they want.

[Mivo]

I mean... I don't mind supporting it because it is a good idea, but why should I play there? I bet I am not alone asking this.

For me, it would be the fact that it will (hopefully) have an international, English-speaking community and run on my iPad. Except for minor issues, I'm perfectly happy with KGS and think it's by far the best Go server so far, but I can only use it when I'm on my PC. I much prefer using the tablet for studying, watching games and, ideally, playing them. WBaduk has an iOS client, which is pretty solid and sufficient for kibitzing, though the community is inaccessible and therefore non-existent to me.

Now ideally, I'd prefer wms to open up the KGS code, or at least the protocol, to allow the development of third-party clients that work on non-Java platforms. But nothing he said so far indicates that he'd even think about this, so Kaya is the next best option, if it gets past the planning stages and gets accepted by the community. As you said, Bantari, people tend to go where other people are, and right now that's KGS.

[danielm]

I mean... I don't mind supporting it because it is a good idea, but why should I play there? I bet I am not alone asking this.
Go servers are like bars. Nobody wants to sit in a bar alone, so people go where people are. But how to get the initial people in?

It is weird, the answer to this question is so clear in my head, but it is not easy to put it into words. I would probably sum it up with one word: Accessibility.

To mention a few thoughts in no particular order:

• As a website, it is much easier to hang out on it than a dedicated application. For example I will certainly have the website open in a tab most of the time, just like currently I usually have OGS in the background. I don't generally have KGS running when I am not playing, especially not when I am doing something else / working.
  
• A modern "web 2.0 style" web application generally has a simple, lightweight, and inviting interface (and everything suggests that Kaya.gs will be no exception). Unfortunately this is not something seen in current Go websites and even less in Go applications, so it may be hard to recognise for everyone how much this is going to lower the barrier to entry.
  
• If there are not enough people yet for realtime games, one can start turn based games if/when that is supported. Something that I haven't seen mentioned yet is that Kaya.gs also has the potential to be a much more sophisticated alternative to OGS and DGS, which is already quite something. Personally, as much as I appreciate the service offered by OGS, I can't wait to switch to something slightly better designed.
  
• Kaya.gs is likely to be a good choice to use for showcasing games for example. As people link to games or other content on the server, they will naturally grow the population of it. There is a convergence of services which is simply not possible within a traditional desktop client.
  
• Gabriel and Patricio seem to be quite good at generating enthusiasm, as the flurry of donations indicates. Many of us will be eager to spend a lot of time on the server/website right from the start. And giving that they want to turn this into their livelihood, they will be aware that active marketing is a requirement.

Sure, it will take a while for numbers to grow significantly. But the early days will also be the most exciting, and that should easily make up for a lower number of players. It's not like you have to stop playing on every other server.

[snorri]

Go servers are like bars. Nobody wants to sit in a bar alone, so people go where people are. But how to get the initial people in?

That's their challenge, of course, but many players will try it out. Think of all the people on L19 who jump through a lot of hoops to play on Tygem or WBaduk. These used to not even have English clients, it was hard to get accounts with registration being broken for long periods of time on at least one of them, etc. There are a lot of people who are just curious about something new.

If Kaya is going to be a very small number of mouse clicks to log in as a guest and see its features, the barrier to entry will be very low. That alone will not be enough, but it's a good start.

[SpongeBob]

I think the need for a state-of-the-art go server definitely exists. Is there are business case for it? Honestly, I doubt it. (Please, please, kaya.gs: show me that this is not true.)

If the picture of this new server that is drawn here becomes reality, the users will certainly come. To compare it with a bar: if a new and fancy bar opens, you will go there and try it out. Others will do so, too. And if you like it, you will come again.

[oren]

That's their challenge, of course, but many players will try it out. Think of all the people on L19 who jump through a lot of hoops to play on Tygem or WBaduk. These used to not even have English clients, it was hard to get accounts with registration being broken for long periods of time on at least one of them, etc. There are a lot of people who are just curious about something new.

People do jump through hoops but that is to play a much more massive number of stronger players and watch professional games. Kaya will have neither of these at the start. Will it be a massive success or look like eurogoserver does now?

[Mivo]

Will it be a massive success or look like eurogoserver does now?

The problem was already with the name: it implies that the server is only for Europeans. That was one of the reasons why it didn't appeal to me. How many Americans or Asians would feel that the server is also for them? There was also very little in the way of advertising and "hype", something the kaya.gs people are pretty good at. Too good, perhaps.

[palapiku]

I have looked at some online chess servers, like chess.com. They seem to be more or less what Kaya aspires to, web 2.0 apps with everything done through javascript, nice platform-independent UI and a ton of cool features.

All of them require paying for almost everything. This is not surprising, as they're business ventures and have to make a profit. Actual playing might be free (as it apparently will be on Kaya) but otherwise the server does everything possible to make you feel second-class if you're not a paying member.

As an example, on chess.com playing games is free, but you have to be a paying member to access your own old games from the archive, or to look up other players' stats.

Now I'm actually worried that Kaya might succeed and KGS will dry out, and then I pretty much will have to join Kaya and pay for all the basic stuff that's free on KGS. I feel like I'm more happy with the status quo than with something like chess.com.

[pel]

The really different thing I have seen is the embedded (streaming?) video capability. But videos are easy these days, and if this is to be the differentiator, pretty soon each server will have streaming video. Plus - videos have an overhead - somebody needs to shoot them. And for that to work, the shooting needs to be consistent and dependable, which usually means money involved, contracts, sending people to events, etc. If nobody at the next Congress videotapes the top games, this capability is meaningless.

Actually i agree. Between monday and today i got several different video embeds ,and in our develompent server you can chat and view videos. You should ask all the other servers why they didnt do this, when the technology to make it easy was available for what..almost a decade?

Maybe you know something I don't, but I can think of a reason why this is not an easy nut to crack. Bandwidth. Video is not easy to scale.

• One to one is easy, you can use xmpp (just an example) for instance to connect both video and information streams. But some users might run into trouble.
  I remember playing on KGS (no audio) while in China for instance, stable/fast enough connections are not easy to come by. This would probably make this set of users feel left out.
• One to few is probably also pretty easy to do using xmpp (again, just an example). Now the problem with video and game-data synchronization starts to be troublesome as well. How do you give all the listeners the same experience?
• One to many can get you into serious bandwidth trouble fast. I'd probably go with some sort of professional broadcasting service here, but then you would get into trouble with giving all the listeners the same experience as you lose control over your video stream endpoints. And a professional broadcasting service is going to cost you money. I guess you could go with some "free alternatives" but that would mean you lose control over the media quality.

I'd say that, in general, the reasons for not having done this before have been either that they haven't thought of it or that they care a lot about the user experience. For instance I bet WMS have thought of it. He has implemented audio support in cgoban using (probably) a speex-codec to be able to reduce his need for bandwidth server side and to give the users a similar experience weather they have good or bad bandwidth conditions.

Damn I sound so negative
I am looking forward your go server though!

[heather]

Maybe you know something I don't, but I can think of a reason why this is not an easy nut to crack. Bandwidth. Video is not easy to scale.

Exactly. One 4 Mb video on a site that gets ~1000 uniques a day will chew through more bandwidth than hundreds of html pages. The benefits endusers can derive from video rarely make it worth the extra bandwidth for webmasters to have to shoulder. Some hosting companies actually charge based on how much bandwidth your site consumes, which would cause a noticeable jump in expenses if a site that didn't previously offer video began to.

I've experimented some with hosting videos on my own sites in the past (as you've probably guessed), and my conclusions have always been that it isn't worth the effort. YMMV; I've never run a Go server. But, just guessing, I'd assume that it will eat up a ton of bandwidth without providing any dramatic benefit to endusers.

[nagano]

Maybe you know something I don't, but I can think of a reason why this is not an easy nut to crack. Bandwidth. Video is not easy to scale.

Exactly. One 4 Mb video on a site that gets ~1000 uniques a day will chew through more bandwidth than hundreds of html pages. The benefits endusers can derive from video rarely make it worth the extra bandwidth for webmasters to have to shoulder. Some hosting companies actually charge based on how much bandwidth your site consumes, which would cause a noticeable jump in expenses if a site that didn't previously offer video began to.

I've experimented some with hosting videos on my own sites in the past (as you've probably guessed), and my conclusions have always been that it isn't worth the effort. YMMV; I've never run a Go server. But, just guessing, I'd assume that it will eat up a ton of bandwidth without providing any dramatic benefit to endusers.

Who said anything about hosting video? A Youtube plugin could be used for videos, and a Skype plugin for live chat.

[pel]

Who said anything about hosting video? A Youtube plugin could be used for videos, and a Skype plugin for live chat.

Well, yes - but

• Youtube will let you upload at most 15min per segment (unless you are allowed to publish more)
• Youtube will not let you do live streaming (there are other services that provide this though like bambuser)
• Using a Skype plugin really does not help the bandwidth problem for the individual users

[heather]

I'm sorry; I might have misunderstood something. If you're just talking about the capability to include YouTube links, then I don't understand how this is even a feature (you should be able to do that anywhere). But you're completely right; everything that I said above argues in favour of hosting videos on YouTube or somewhere else that specialises in such and won't dent your bandwidth. I have no problem with providing video; I only question the soundness of the decision to host video oneself (which may or may not even be a proposed feature here).

[danielm]

There are plenty of services which would offer to host streaming or static video for embedded content (which is not the same as simple links). Whether kaya.gs ends up using such services or host the content themselves, bandwidth is definitely affordable enough these days (otherwise nobody would offer those services either). Of course you have to have a plan to turn your traffic into revenue, otherwise you are screwed either way.