Altering that JRE license to my own ends

EDIT: I’m doing a slight thread-jack here, and ignoring the JRE licensing issues just to look at the issues I’m highlighting in this post. I realised it could look quite confusing without this explanation :slight_smile:

Because the value of letting all your partners (i.e. people trying to develop games in java) know what’s happening outweighs the probably tiny advantages you give to competitors. AFAICS this is the fundamental equation that justifies Sun supporting java games developers publically instead of behind closed doors.

As a matter of interest, I would love to know what competitors you feel are going to benefit by any non-trivial amount to know roadmaps etc for Sun’s involvement in java games development. I personally can’t think of any examples right now?

You cannot support java games development with that attitude; this is probably why people like myself feel you are not managing to properly support the community and that you are damaging the community - this attitude would certainly prevent you from doing these things.

In answer to your specific examples, what most good marketing companies (like MS) do is to tell specialists and partners in advance roughly what will be announced - and this is usually done in public, but because they don’t send a press release to the news sites etc non-specialists don’t notice.

Then, at the event, you still get your big exciting moment - but you have the benefit that you haven’t caught your partners by surprise, and that when the news people etc go looking for commentary by specialists, they get some good responses because these people have had time to think about it and worked out how great it really is. (PS if you’re running a community like JGO it’s usually a good idea to in fact hand out lots of “quick reference/FAQ” material before the event that doesn’t give much away but contains brief explanations of where the new thing will add value - so that when they’re contacted by press etc, or write editorials in magazines etc, they don’t misinterpret your message).

Just to be clear, what you might expect someone to reveal “roughly” would be a lot more than the GTG usually does - and a lot earlier. For instance, last year you told us not to start a Wiki and various other activities but said you couldn’t tell us why. What you should have done (now that I know what you were keeping secret!) was to say “We will very soon have a lot of extra community features, a special site, dedicated resources within Sun that include lots of community involvement things that let you work together and share info, code, and project planning. The precise nature of all this will be announced during March”. Instead you said nothing, and only belatedly responded to the Wiki attempts by saying please don’t do it it will be wasted effort. You told us after we’d wasted the time thinking about it and starting it etc; you could so easily have told us weeks before.

WHS. And it was just that strategy that caused a lot of friction when the GTG came into being.

As far as the efforts of the GTG are concerned, I am still in exactly the same position I was in 3 years ago when I started out - nothing has changed. And so are all the other developers I talk to in my market segment. Not a one of them wants to use Java, more or less precisely for the reasons I outline. The situation is not improved at all!

Dweep could not have been written in 1.1 Java. First of all it’s fullscreen, which couldn’t be done in 1.1. Secondly it’s only 1.3MB download or something. He’d have had to ship a 5MB VM with the game (and continue to do so to this day) which would quadruple his bottom line costs. Remember he started in 1998, when most of the world was just waking up to 56k modems, and server bandwidth was considerably more expensive. And besides, Dweep is a 6 year old game, and it looks it. You need to give people something better or there’s still no concrete incentive.

I’m going to keep hassling you about this and I won’t stop until I’m dead :slight_smile:

Cas :slight_smile:

How would you characterize a game like Tribal Trouble? The internal web start demo has code, terrain, water, sky, GUI, 2 bitmap fonts, a single unit (out of 6 in the final game), and a single building (out of 4 in the final game). And guess what? It’s 2.9 megs! Adn we intend to keep the game that way (preferably < 10 megs) and distribute it through the internet, so we pretty much share most of Cas’ griefs.

Bottom line: this is not a problem converning developer! The topic title is probably just Cas’ cocky way of attracting attention…

  • elias

[quote]He’d have had to ship a 5MB VM with the game (and continue to do so to this day)…
[/quote]
I think Chris’ point was that if it were coded to Java 1.1 that shipping a VM would not be necessary since 1.1 VMs were on most of the target machines.

I see your point about not much changing form your perspective - since you are doing LWJGL (hosted on SourceForge, not java.net) independently from the ‘Core APIs’. But to others the GTG’s efforts in establishing the Core APIs and open sourcing them in an attempt to please more people by having ‘Sun supported’ gaming APIs yet allowing development to move at a faster pace than could be done through the JCP (as JSR-134 was cancelled partly for this reason I believe) are welcomed changes. So I would give the GTG credit where it is due.

That doesn’t mean that we shouldn’t be voicing our concerns for areas where the GTG could improve. I agree with blah, that the communication level can be improved without spoiling the surprise. Perhaps key players at the indie level could sign some NDAs as well to become more involved. I’m sure Cas has would have no problem signing an NDA if you included him in the discussions about open sourcing the core APIs - his experience with LWJGL would likely have been valuable.

Again we are aware of small hints of what will be announced at GDG… Jeff mentioned work on some server-side technology… I hope that there are some industry partners that have signed the NDA for that one.

…if we’re anything to go by, not many. I wanted to sign that NDA more than twelve months ago!!! (I can probably dig out copies of the emails concerned). Back near the start of January, I once again brought it up with Chris and he said sure - but again nothing happened.

And we’re one of the few companies that really know about this stuff - going by my research of the last almost 6 years, we’re in the top 10 companies worldwide in terms of understanding the hard end of server-side game development (i.e. discounting all those with experience of lobby games, battle.net, yahoo games, etc). That’s without looking at all the people I know personally who specialise in this area, and what I know of their past successes and failures, and what they need and look for on future projects. Surely something of some value there in terms of partnership?

At the moment, my current expectation is that Jeff’s going to reveal something at the GDC that is in direct competition with at least some of our commercial activities, and that the GTG has known this would be the case for some time but decided to work against us rather than with us. I’m not being paranoid - look at the history and circumstances surrounding JInput. Then bear in mind the extent to which we’ve not even been allowed to see the NDA that so many times now we’ve been told we could/should/must sign.

Shrug. So, we get on with our own work, and just have to assume the worst - that in March something bad will happen - when we could instead have been sharing tech, experience, and skills with Sun. If anyone has an explanation of why this is the status quo, I’d love to hear it; it seems a weird situation to me, and yet also seems not uncommon at the moment.

Ok, so this thread is way off topic, so this will be the last time I post off topic.

So ok. If you want to formalize as “partners” then are you suggesting a partners program to distribute information that, while not under nda, but is otherwise sensitive?

I am still at a loss in understanding how you think we have not supported this community? Community forums? Code repository? Access to APIs? Regular participation in the community? If by lack of support you mean that we can’t address EVERY single concern of the entire community then is that a reasonable request?

[quote] As a matter of interest, I would love to know what competitors you feel are going to benefit by any non-trivial amount to know roadmaps etc for Sun’s involvement in java games development. I personally can’t think of any examples right now?
[/quote]
To be blunt, if we believe that there is a business reason to not disclose what we are working on, to a public forum, that is a business decision we have made and stand by it. Would you be willing to share with the community all of the partners you are working with, details about the techology you have in development, and provide them to a mechanism that has no controls around them? Let everyone see who and what you are specifically targeting so that they can approach them as well. I dare say that Sun Microsystems would have much more at stake in this scenario then GrexGames would;)

[quote]In answer to your specific examples, what most good marketing companies (like MS) do is to tell specialists and partners in advance roughly what will be announced - and this is usually done in public, but because they don’t send a press release to the news sites etc non-specialists don’t notice.
[/quote]
Incorrect. Flat out incorrect. MS and anyone else that is revealing information to partners that may be sensitive do so under non disclosure and most definately not in public. Work in a systems channel long enough and you will see what I mean.

[quote](PS if you’re running a community like JGO it’s usually a good idea to in fact hand out lots of “quick reference/FAQ” material before the event that doesn’t give much away but contains brief explanations of where the new thing will add value - so that when they’re contacted by press etc, or write editorials in magazines etc, they don’t misinterpret your message).
[/quote]
Good point. We will put together a general “pre GDC” document for this community as we get closer to GDC (See, I’m not all no,no,no :))

[quote]Just to be clear, what you might expect someone to reveal “roughly” would be a lot more than the GTG usually does - and a lot earlier. For instance, last year you told us not to start a Wiki and various other activities but said you couldn’t tell us why. What you should have done (now that I know what you were keeping secret!) was to say “We will very soon have a lot of extra community features, a special site, dedicated resources within Sun that include lots of community involvement things that let you work together and share info, code, and project planning. The precise nature of all this will be announced during March”.
[/quote]
Good point. I will work to better communicate our intent moving forward.

Listen all, it is important to understand that WE DO appreciate this community and all of the effort you put into it. I truly believe that we are all here because we have the same passion to see Java technology succeed in game development, from indies to professionals. I also think the fact that we are spending time hashing out these issues shows the strength of the community as well.

Like any good relationship, communication is the key and we will work to more effectivly communicate the GTG’s intent moving forward.

Now that you have a community board, it is time to put them to work and outline how best to do this.

-ChrisM

[quote]How would you characterize a game like Tribal Trouble? The internal web start demo has code, terrain, water, sky, GUI, 2 bitmap fonts, a single unit (out of 6 in the final game), and a single building (out of 4 in the final game). And guess what? It’s 2.9 megs! Adn we intend to keep the game that way (preferably < 10 megs) and distribute it through the internet, so we pretty much share most of Cas’ griefs.

Bottom line: this is not a problem converning developer! The topic title is probably just Cas’ cocky way of attracting attention…

  • elias
    [/quote]
    So, if you had a customer who wanted to play this game do you think a one time download of 12-14MB would deter them? Hell, you buy WinXP today and you need to download over 80MB of patches and updates. It has been so problematic for MS that they are releasing patch CDs to consumers!

-ChrisM

[quote]Ok, so this thread is way off topic, so this will be the last time I post off topic.
[/quote]
Yes, that was my fault.

So I’ve opened a new thread for this topic, and would appreciate it if you’d continue the discussion here:
http://www.java-gaming.org/cgi-bin/JGNetForums/YaBB.cgi?board=suggestions;action=display;num=1077304454;start=0#0

[quote]So, if you had a customer who wanted to play this game do you think a one time download of 12-14MB would deter them?
[/quote]
Yes, Chris, this is exactly what we’re saying, and I’ve got hard-earned proof of it, and also a lot of evidence from other key players in the industry. Please understand that I think that Webstart is tha b0mb, but it only works for some people, and definitely not in this space.

Cas :slight_smile:

[quote]Nearly off topic :wink: I have decided that it would be far better now that Sun do not try and include a binding in J2SE.
[/quote]
Hm, I’d really love to see a J2SE with OpenGL/3d binding. But since it’s nearly off-topic I won’t continue dreaming.

[quote]Back on topic: I am exactly saying that Java is competing for mindshare in companies like Dexterity to produce “casual” games. But I am not talking about just casual cames - I’m talking about indie games, which range from the very simplistic Dweep Gold and Fitznik right the way up to the fantastically presented Hamsterball or Starscape Moonpod. (…) The only thing all the various indie games have in common currently is that they are of downloadable size and that they are extremely simple to install and run. Download size kills this market. Complexity kills this market. Even reliance on OpenGL drivers is detrimental although not nearly as bad as some indies think (a lot of them are still deliberately working to DX5 level to capture more customers).
[/quote]
OK, we talk about independent developers then. Their games range from Blockout likes to Moonpod likes.

I think you dig in your own points with your above paragraph(s): while for Hamsterball the 6,5 MB size is indeed half of the current JRE, for Moonpood sized 35 MB (optionally deployed on CD) the JRE’s size doesn’t matter. Customers won’t care if they download 35 MB or 45 MB (*). Let alone when the game comes on CD. Add better graphics/sound/data/bla to any game and/or more complexity and the less 10 MB for an initial JRE matter. With any additional Java game being on the market people will potentially just download the JRE once. Add to this that the majority of new PCs are being shipped with an JRE: I can’t see your problem in the order of magnitude you describe it.

Plus: we’re just talking about Win32 “indie games”. Wasn’t it you who said that now 50% (?) of your sold games go to Mac users? Do they need to install any JRE?

To “easy of installation”: Why should it be more complex to start or install a Java app/game than a native game?

[quote]I know full well that Sun has sights only on the big studios like EAGames and Take2 and Sony. Otherwise I’d see a product roadmap outlining how Java was going to help the indie market sector. Currently there is no place for Java in this massively expanding industry sector except if you buy Jet. Sun are simply ignoring a marketing opportunity.
[/quote]
Well, supporting independent developers is good. As Chris said: SUN does do it already. Look at the many help and ideas and discussions the SUN people bring to this forum, to the gaming API projects etc. Who pays for this? SUN and/or the guy’s sparetime. I am thankful for their help.

There are so few Java games no matter indie or fullprice that you can say: there are as many independent game developers using Java as there are fullprice game developers using Java.
If game developers love Java because they’re more productive with it, they’ll use it. If your game is superb you have the power to influence your potential customers. The fact that most game developers don’t use Java (indie or fullprice) is hardly being influenced by the JRE size. The main reason is: there are hardly professional game developers knowing Java or thinking Java would do the job. That’s the problem. We’re attacking Don Quichotte’s windmills in this thread…

(*) If you bundle the JRE 1.42 with your game. Plus use a slightly better packer than zip, JRE takes 10 MB instead of 14 MB. You even squeezed the JRE more with another packer.

No, unfortunately that’s not entirely true.

The ATI Rage (pro) chipset is just a pile of s-word :slight_smile:

ID used special shaders in Q3 to bypass these limitations.


@topic (on of em)

I don’t fully agree that a trimmed down version of the JRE is the best solution, because it leads to more overhead if the user installs several games.

What we really need is that what Cas already suggested several times. Native webstart (~1mb maybe) and a modularized JRE (1-3mb chunks).

I say this time and time again in various different guises: users care 10x less for overhead than they care for complexity, but there is a threshold that cannot be exceeded for either.

The first rule of thumb is that the number of sales decreases by a factor of 100 for every additional installation step required. Yes, 100. This I have determined from my own controversially collected statistics and analysis of over 20,000 installations.

The second rule of thumb is that the number of downloads decreases exponentially with size but follows some interesting human characteristics that cause it to quantize. That is, there’s a shallow curve between 1-5MB, followed by a steep drop and a shallow curve up to 10MB, followed by a steep drop and a shallow curve up to 100MB, followed by a steep drop and a shallow curve thereafter to ISO sized distributions.

Again, I’m asking folk who don’t want to believe me to go and talk with the actual pros that are in this business. Head on over to www.dexterity.com/forums and register and start asking questions about Java and runtimes and size and complexity and so on. There’s no point me repeating myself. You need grey hairs and acumen along with your proof. Go now!

Cas :slight_smile:

[quote]Where do you see Sun just lumping Java technologies together and creating useless profiles?
[/quote]
Sorry, weighing in a little late on this one, but I just noticed this and want to pick up on that comment. While Sun hasn’t (from my point of view) ‘lumped Java technologies together to create useless profiles’, they have seen fit to abuse the Java brand with the ‘Java Desktop System’ which I believe is a significantly worse offence. Only my opinion of course.

Yeah, I know I left my self open a little bit on this one. The difference is that the Java Desktop System is not labled as a Java technology, nor has Sun ever said it was. It uses Java technology but is not a “Java ™ Technology”. A bit different.

-ChrisM

Looks at latest marketing email from Sun … doesn’t say it isn’t Java technology either. The two sentences containing the word Java are:

[quote]Simplify software deployment with the Java[tm] Enterprise System
[/quote]

[quote]A Sun Java Desktop System evaluation CD
[/quote]
There is no qualification anywhere in the email to suggest what Java means in this context.

Rather like Windows Server.NET really, and Microsoft later realised that was a mistake.

Yes no maybe so. I understand your particular concern - one of being able to have a downloadable in the smallest possible size to attract customers for smaller click-through content types. This has been a discussion topic within my circles, particularly when it comes to a Java vs Flash scenario. However, the business development folks are okay with ‘reasonably large’ content for games. They have come to accept it as part of the “downloadable games” environment.

It would be nice if the download of the content platform wasn’t a larger download than my content, but it isn’t the largest of my concerns at the moment - different motives and content types. I can see how this is a large problem for you however. I also see why it comes up from time to time in my meetings as well.

Am I the only one who feels a bit uncomfortable about trusting the JVM that the user has installed to perform correctly?

I’m in the early planning stages for a small project that I’ll either build in Flash or Java. Flash offers better media authoring, great support for both PC and Mac, but costs money. Java is free and I might need more advanced text and sound handling than I think may be possible or desirable to build in ActionScript (really depending on what kind of music I can get my hands on).

There is a good chance that my target users will be tech illiterates, or at least tech-trouble-dont-have-patience-fors, and my experience from releases at different Java companies in the telecom industry tells me that it is a very bad idea to release anything that hasn’t been tested on the specific JRE version and platform.

I don’t want to make the user go through the trouble of installing a JRE in addition to my app, especially if she already has a later version installed (one for which I haven’t tested my app). I might actually just run the app directly from the CD without any installation whatsoever, so I will most likely just include a private JRE version with my app.

A smaller game VM would decrease the size of my app, but I’m not sure that will be a concern since I don’t know if I’m going to support internet distribution yet. I want to stay away from direct installations through the web browser as I think it might be important for my app to offer an experience that doesn’t connote “free hobbyist web-creation” (or worse, typical 1998’s Java applet).

WebStart accessing a CD-rom or the network outside the browser could be an option for JRE installation since it allows me to target specific versions, but I don’t really trust it as it doesn’t work on my own computer (I’ve got JDK1.4.2 and others installed). I get a message that says “Bad installation. No JRE found in the configuration file.” when I click a webstart app link (message is localized so exact wording might differ in English). I know I can probably fix it, but I’m not sure that also applies to my target user if she gets the same problem.

Edit: The point of my post is that configuration management is another strong reason for embedding a JVM with the app, and it would make sense for it to only contain the necesseties (like a game VM).

Another technology that requires constant downloads is QuickTime. And Real for that matter. Is it more difficult for people to dl these? Again, I really understand the size concern being voiced by some, but agree with GP that the business rational behind it is becoming less and less important.

-ChrisM

[quote]Another technology that requires constant downloads is QuickTime. And Real for that matter. Is it more difficult for people to dl these? Again, I really understand the size concern being voiced by some, but agree with GP that the business rational behind it is becoming less and less important.

-ChrisM
[/quote]
The important thing to note is that the concern tends to vary from region to region. In the United States, there are certain things that we can count on from a fixed cost perspective and broadband adoption perspective. Depending on the stats of the European stomping grounds (of which I have no hard demographics) that Cas operates in, he may see an entirely different picture. Much of what we’ve discussed here is a very US centric approach to content distribution for a particular demographic. This demographic tends to be web savy, make a certain $$ per year, and not have any particular concern with respect to download time. That’s my particular business case. Our economies of scale are going to be a little different that Cas’. It is entirely possible that from an adoption and fixed cost perspective, Cas’ content may be a lot more expensive when all things are considered. First off, Cas’ content isn’t free for end users. He doesn’t have a content pipeline which is making regular money to subsidize his fixed costs to decrease the development costs and therefore the expected sales price of his content. As I’ve been working with some folks in an international perspective I have come to understand and respect their concerns. Not saying that you don’t Chris - just that the business structure under which we operate as (practically) infinitely larger and heavier capitalized companies may be impractical from his perspective.

Clearly there is nothing that the GTG can do to assist him in that particular manner because you don’t control the licensing or IP departments. However, it would be nice (and one of the things that I mentioned in my bid for a board seat) to take Cas under our collective wing AS a business case and help him get his content out so we can clearly understand that indie business, how to help it, and what is the reasonable road map that people in his position can follow towards success. Clearly, the java.com angle is an important one. We should gather demographics from that. The midlet sales area is another one that we should gather information from. While this information may not be publicly available at this time, there is no reason that it couldn’t be shared with the 1 or 2 companies,projects that we take under our wing. This isn’t to say that they would be guaranteed sales or success or anything - but we need to take one of these through the cycle and see what all of the issues are from start (development tool chain) to finish (someone buying content) and see the entire picture and how we could assist others in this manner. When I think of ‘leadership’, this is the type of thing that I think of. This is something within the GTG charter (because in order for Sun to turn a profit you’ll have to understand how someone can capitalize in this space), and is in the best interest of both Sun, its partners, and the overall java gaming industry.

While there are other projects out there which are games and such, I believe PuppyGames is best positioned to take to the ‘next level’ and see how far we can take it.

[quote] the business rational behind it is becoming less and less important.
[/quote]
myopic vision I’m afraid. we’ve got a similar problem here in NZ, where those in Auckland (the largest city) tend to forget that there’s anything outside of Auckland. It’s the same in the US. Where they tend to forget there’s anything important outside the borders of their country.

Anyone who thinks that download size is become less an issue is probably based in an area where broadband is widely available and relatively cheap. Coming from a country where we have a monopoly telco in control of the broadband loop, and where broadband prices are therefore… unreasonable… a majority of internet users are still dialup based. And I know the same is true for a lot of other places.

If you’re targeting the US, then download size is of lesser importance, if you’re aiming at another country, or internationally, then of course it’s going to be an issue. If you just want to target the broadest userbase possible, then it’s still an issue.