Start Fresh?

[quote]- given lwjgl and jgo are both relatively low traffic, it’d make sense to merge them together in order to concentrate everything in one place and avoid dispersion. Also this mean, theoretical, lower maintenance… two birds one stone. Apostolos and Spasi seem to agree and share the same point of view
[/quote]
Nice offer and well worth considering!

A consideration: jgo could also be a sort of hub and could refer lwjgl questions to lwjgl? I am thinking along these lines because there are a number of java-based libraries, e.g., jmonkey, jogl, others?

[quote]- @philfrei, why don’t you join us on slack? we could coordinate us faster and easier
[/quote]
Do you have the link? I’ve dropped in a couple times, but don’t have the link handy and my “working memory” for these kind of things is old and creaky.

Slack allows quicker turnaround, but it’s worth considering that this thread is a visible public record. I don’t think that there is a particular need for speed, even if the natural tendency is to experience impatience. If you are going forward with setting up a Discourse version of JGO to test, that would be great! Hopefully I’ll get over the tech hurdle with JForum2 and we can get these both up and running for comparison purposes before too long.

Good news on my end: Ulf Dittmer, the author of JForum2 is being VERY responsive and helpful, has built a Jetty-enabled version which I just downloaded and am trying out. Progress can be seen on this thread.

[quote]I think it would be better to take some working software instead of preffering Java-based forum software.
This forum is about java-gaming and not about java-forums Smiley
I’d prefer to have staable and efficent forum against java-based forum.
[/quote]
I wholly agree with this. But I think it is to soon to say that JForum2 is “not working” or unstable or inefficient. What is lacking is my knowledge and experience with installing servers!!

Thanks!

If you ask me, that’d be awesome, this would help focusing efforts, precious resources and valuable members in one unique place.

However their corresponding authors have to agree for that. I know for example jmonkey has already their forum (which is exactly Discourse, so maybe this may give more possibilities for a potential fusion). Regarding jogl, I assume it was (long time ago) part of jgo, but then there was a divorce because of flames and either sven and gouessej, as far as I remember, are against it and they prefer to keep their own separate forum and discussions.

Anyway I find the hub idea absolutely fantastic, it may greatly help to revitalize the java gaming/3d-graphics community

Sure, https://slackin-bfxneqmzsp.now.sh/

Yep, I agree, that’s why I wrote that small update following the developing on slack

Nice to know

See you on slack :smiley:

(hmm notice how snappy and quick his JForum site is…)

Cas :slight_smile:

So it seems you are actually trying to pursue this. Alright; might as well add more opinions then:

When looking for a new forum software, do not take the “anything that works” approach. That is how JGO got stuck with SMF in the first place, and going by what information Riven dropped over the years, it is almost impossible to make any adjustments to it at all.

If you want to do this, don’t go by how it was created (JForum - let’s be honest, you are only considering it because Java) nor by what currently appears popular - popularity, almost by definition, changes.

What are the actual requirements here?

The long load times we have right now (several seconds for me, sometimes a page load just fails outright) are IMO problematic. They make the whole site feel offputting. Please consider speed a factor even if it is not in fact a neccessary feature.

You might want to go with something extensible. Requirements for what a forum should do change over time, and it should be possible for the forum to change with that, if reasonable.

You might also want to use something that is, if case be, more easily migratable. From what I’ve seen here that is a big issue, a good part of why SMF is still in use at all here.

Also, you might want to pay attention to keeping a “look-and-feel” that seems a little more modern; at the same time, using just the newest trend is bound to end up being the most recently outdated trend soon.

After all that, maybe start small? No offense, but I am not getting the impression that anyone involved here is really fully knowing what they are doing.

Before you start talking about things like dismantling several established community sites to try and glue them together somehow, please consider to just start small and see how it goes.

And if you somehow get all that settled, there is then still the very important question as to exactly how you want to go about migrating all the people.

Thanks for your hard work guys.

@Riven: Would be interesting to get your opinions, as the guardian and generous caring uncle of this forum for so long.

[quote]If you want to do this, don’t go by how it was created (JForum - let’s be honest, you are only considering it because Java) nor by what currently appears popular - popularity, almost by definition, changes.

You might want to go with something extensible. Requirements for what a forum should do change over time, and it should be possible for the forum to change with that, if reasonable.
[/quote]
2cents: Some good points, but using Java might actually be the best way to keep the forum extensible and migrate-able since we know how Java works. Nobody wants to touch the SMF forums since programming in PHP is a nightmare. For this reason I think that JForum2 would be a good solution.

By the way, JavaMelody is a popular tool for tracking memory usage and other things on java web servers and has some interesting usage stats here:
http://javamelody.org/usage/stats

JForum2 looks old, and JGO looks more modern then JForum2.

I think if we are going to move to a new forum software, you should move to a modern design with a good responsive / reactive web site. A lot of people use their phones, make sure it supports mobile devices well, and not just “it works” because, because I can navigate JGO on my phone already, just not ideally ( having to zoom in to make sure i click the right link ).

It doesn’t really matter what language it is written in, how often are we going to make unique changes to the website?Which will potentially cause us to not be able to update the software. If we can’t update the forum software, we are going to get stuck and unable to update ( which I believe has happened to this forum? ).

Also, trying to grab the attention of new users, having an old looking forum could spark the idea of the website may not be active.

I totally agree about starting small.

However with a 1GB, 1 CPU, 25 GB @5$/mo you cant start smaller than that :smiley:

the hub purpouse is just an… idea. It’s of course on the long run. I think nobody actually believed to apply that in few hours/couple of days.

You start small indeed, by sharing your goal with other people, see what’s their feedbacks and reactions, contacting even more people, etc etc… and then step by step, if it makes sense, you go forward.

Start small on the goals, not on the means. I’ll assume judging by the rest of the post you understood me correctly already though.

Before moving or changing JGO it would be useful to look at what this forum is, to who and clearly state the goals.

Its ok to say software x is mobile friendly and this framework is new and exciting but will it fullfill the needs of JGO and its members? This is quite a small community and while change could bring more users in its possible if the change is managed badly then you kill it off for good.

Change has to be planned and exectued well, you want to retain the core users, add value to the site and attract new members.

If you do start fresh it would be nice to leave this intact and maybe create archive.java-gaming.org.

  • thinking about all the links already pointing here that would break, a new forum should probably use a new path to leave all old links live *

Any thoughts on how we might engage in this process? Can it even be done on a forum? Maybe on a wiki page (where everyone has a chance to make edits)? Or maybe a moderated thread, where people give input, but the moderator updates the first post (with the goals) according to the replies that come in?

It would also be good to hear any particular concerns or worries that you have, as well as goals that you want to make sure are included.

This is far from finalized, but one idea is that the new site be given javagaming.org, and the old stay at java-gaming.org. We have both. I’m thinking all the topics on the new could have a pinned post pointing to their counterpart topic on the archive.

Wouldnt it best the other way round? Or does it have to do with the fact that this would break all the external references existing so far?

I’m thinking that maybe we could have the url java-gaming.org send people to javagaming.org (new forum), but have pinned links in javagaming.org that allow people to easily get to the SMF forum. Just an idea. But that would preserve all the external links, right?

I’ve gotten JForum2 working on my laptop. I’m starting to learn how to customize it. There are some easy things I’ve done, like setting up categories and forums, and putting in pinned posts for links to the existing JGO. Less obvious (to me, being a newbie at this) are things like getting the logo right (will have to explode the war and see about dropping in a copy of our current, but it would be nice to know if there is a way to specify the link), or getting a “child board” type of functionality (maybe involves making another template that can be designated for the “forum” category?).

There are instructions for getting “developer” access. I’m guessing a lot can be done first to customize without tweaking the Java JForum2 code, and instead putting in or editing material in the templates subfolders of the project.

I suspect (am not sure) that after populating with some users and posts, it will be possible to inspect the underlying database and compare it to the database built by the SQL code from last year’s dump. Whether we use that to bring over all the data or to (more likely) pick and choose aspects remains to be seen.

Anyone else want to explore this? I’m happy to let you know what I did to get this set up on my laptop. It wasn’t hard at all, once I switched to using PostgreSQL instead of MySQL. It would be great to have someone else to work with to learn about forum admin.

(I suspect that the MySQL problem was something dumb like using characters that the code treats as escape or special chars in the password. But a lot of people prefer PostgreSQL to MySQL so maybe I’ll just let that go.)

It could well be that Discourse has a lot more configuration built into a GUI. We will eventually get both up for comparison and a vote, I assume.

So what happened here ? did Riven die or something ? I’ve been away for a little while…

Hey guys!

Sorry for taking a long gap. I got involved in a road accident recently which could have been a major one and luckily made it out with a minor scratch and a ligament tear. Though it was a minor one, I decided that I spend some time with my family.

There are multiple tasks upon me now that I’m back:

[ol]

  • Get back to my work and carry out some backlogs.
  • I’ve started moving my website away from GoDaddy to a DigitalOcean droplet. I have to finish that.
  • Get back to this task of migration.
    [/ol]

I’d really like to try it out myself, so we’re starting a forum instance using Flarum for my SilenceEngine. Am planning to finish this by next week. Will let you know guys soon.

@philfrei Sorry to say this, but my old SMTP mail server is now dead. I haven’t received any mail after sending my last one since it is dead. If there is anything, please send me again via PMs. Will let you know once it is up.

I’ll happily change the DNS config of java-gaming.org to whoever shows he/she can use the earlier provided anonymized JGO-dump and turn it into something viewable, preferably retaining the current URLs for every thread/post (this requires quite some trickery) so that inbound-hyperlinks continue to work. Whether that’s in read-only mode or not, doesn’t really matter. This requires moderate skills in hosting and SQL (unless you find some conversion-script). It doesn’t even have to run on any VPS. If it runs on your laptop, that’s perfectly fine. You can (ssh)tunnel your local service (or config your router) and make it reachable from the internet without much difficulty - which saves you $5/month in the trial-and-error phase. Having JGO running on a local system significantly eases development on it. You don’t want any change to be put on production, odds are you’ll break something every other day.

Please realize that ‘infra’ (getting a VPS, configuring it) is a tiny fraction of the effort.

As for the domain name javagaming.org (not java-gaming.org) - that is owned by Oracle, for which ChrisM has (had?) the contacts to change the IP - this might be a legally complex and/or introduce delays at the worst of times, hence I opted to go with java-gaming.org.

If you wish to leave JGO as is (disable the signup/login pages), and make a subdomain to host the new site, or any ‘folder’ (like java-gaming.org/forums), then the same applies to what I mentioned about creating an imported/migrated read-only version of the current JGO: get something working, get feedback from the community, share your proof-of-concept, pretty much this means you have to show you’re capable of handling this. We’ll move from there.

Good luck!

I’m feeling discouraged about my ability to help here.
The tech described by Riven is over my head. I don’t understand a lot of what he is describing.
I made a pretty big push to get enough server basics to install Jetty and JForum2, but seem to have mucked up the working install of the latter by tinkering with (and I thought correctly restoring) a configuration file made by @princec.
My research on MySQL “batching” doesn’t seem to have proved to be much help for SHC’s efforts either.

:frowning:

Semi-related–decline of JGO site also due to fact that more and more tech is moving away from Java? The example I ran into today came from looking into Chromebooks. In order to run Java programs on them, it seems to me the programs have to either be cloud-based (java server & .war running HTML/CSS/Javascript for the GUI, via FreeMarker? is that the main path for java to run on the web?) or maybe the java programs can run with Crostini enabled? I think one can install OpenJDK 11 if Crostini is working and make a program with an embedded jlink-built JRE with that. Either case, not the easiest choice, and not the best business scenario if you have to get customers to install something (e.g., Crostini) in order to run your software.

EDIT: Looks like Crostini and Debian support has become “standard” as of Chrome OS 69. https://www.androidcentral.com/how-install-linux-apps-your-chromebook
I know Debian has OpenJDK 12, I just installed it on the Ubuntu server.

“decline of JGO site also due to fact that more and more tech is moving away from Java” :

Yes, MAINLY due to fact that more and more tech is moving away from Java.

In my company, I used to program in Java from 2006 to 2018, as everyone there was doing Java, but now I have switched back to C++, because only our legacy projects are now made in java. (And because it is much simpler to program in C++ now that in 2000)

The hype for Java is mainly over.

The getting the existing links to work is indeed a trickery here, as we have two cases:

First is to change internal links. Links to other posts should be replaced with new links automatically. I have been working on a post transformer that automatically does this. This one is simple indeed.

Second is to change all the inbound links. I’m not sure how fine are you with giving out 301 for those links. If you are fine (it will impact SEO, but I still prefer this one), then I can generate a script that does all these redirects when the porting happens.

If you are not fine with 301 redirects, I’ll have to add another extension to Flarum that dynamically rewrites these URLs. Though both of them works, I prefer the former, give out 301 with the new URL.

@philfrei: I have read your message, but was wondering how to respond to it. I’m fine with any case, whatever forum software we use, the ultimate goal is to see how well it survives in the future. However, personally, JForum2 doesn’t appeal much to me. The main reason is because of the old UI it possesses. And I don’t think it is being maintained either: https://github.com/rafaelsteil/jforum2

By the way, please don’t worry. The research you made could be helpful to me. Truth as you already might know is that I haven’t tried it. I was waiting for the extension APIs of Flarum to get stabilized. This was also a learning experience for me, cause I never previously worked on databases. I will continue to work on this.

@SHC The JForum you linked is obsolete. But there is a more current version https://sourceforge.net/projects/jforum2/ which is being actively maintained. The UI is editable HTML/CSS/JS, interfacing with Java using Apache FreeMarker. For support, there are forums at JavaRanch (which now uses this updated version) and at the developers’ own forum here. They have been encouraging of our efforts and very responsive, even making two updates specifically to address issues I brought up.

The new JForum-2.6.x works with MySQL, PostgreSQL, or can be run with an embedded HSQLDB. But as far as how we link up old/new posts and preserve links, I am going to defer to others. My skill set is not advanced enough.