The Last Homely House
General => The Golden Hall => Topic started by: Kralik on June 28, 2020, 11:37:02 AM
-
For the next couple months I'm going to make an effort to tidy things up with this website and fix outstanding issues. My biggest priority is going to be upgrading it to HTTPS, which will require upgrading the forums and Wiki software. Unfortunately both have a lot of hacked source code changes, so there will be a time when things will be partially working or not working during the transition. For example, all of the automatic card image linking is custom code and since it's been over a decade since I wrote it, I hardly know how it works!
We might lose some custom mods like the gold coins (does anyone use these anymore?). Those were an attempt to replicate a cool feature from Cobra Cards where users could "buy" forum upgrades, but I was never able to get it to work completely. Linked images that aren't on HTTPS sites will also break but that may be a minor issue.
In any case, now would be a good time to let me know if there's anything that's not working well with the Wiki. Does editing work? Can new images or articles be added? I know the card text isn't editable (by design), but are there any glaring errors? Please feel free to PM me any pertinent details or email me if something urgent comes up.
No guarantees of success but now is the time if it's ever going to happen! :mrgreen: \:D/
-
Oh man, the wiki.
So, first of all, let me just say that if upgrades are a thing you're considering, I would highly suggest just installing MediaWiki. It's the software Wikipedia uses and I've used it for personal projects and other game wikis and it's far more fully featured than dokuwiki. If we install cargo or SemanticMediaWiki, we can even make pages that query other pages, such as the Hero page on the Artifact wiki I worked on here (https://artifactwiki.com/wiki/Hero). That big ol table is the sort of thing we have to do manually here, but is like ten lines of query code (https://artifactwiki.com/index.php?title=Hero&action=edit§ion=2) all in the page itself, which automatically updates in the event a new Hero page is created (note the broken rows at the top where someone hasn't filled in the proper info on that hero page, lol).
I've also written bots that interface with MediaWiki's API, which means that mass edits would not rely on your access to the base file system and could be ran by the community (assuming you authorized the bot initially, that is). Migrating would not be difficult in that case.
Anyway. If you're not willing to do that, here's the list of things broken in the current wiki off the top of my head:
- Cards can't be added, nor edited (as you noted).
We recently got all the deets and images for the 30-some-odd W cards out of LOTRO and those can't be added without your help. While working on that, I lined up the LOTRO and TLHH data in a database and produced, oh, about 30 discrepancies in wording, typos, etc. I couldn't fix any of those so long as you're gone.
Frankly we should also be able to add the Hobbit draft set that was worked on, as well as other fully released virtual sets (including potential future PC-ran virtual sets coming up) without needing admin approval. A card *should* just be a plain ol' article, but because it has to interface with your DB in some special way it's not possible for the community to organize this.
- Uploading files is hit and miss. In the past this has been due to running out of hard drive space, but at other times there have been file size limits or other random unexplained problems that have lead to half the rulebooks just being hosted on my Google Drive (http://lotrtcgwiki.com/wiki/rulebooks) since I couldn't get the upload to work. Granted I haven't tried recently, but I had no reason to believe it spontaneously started working.
- SVG support would be good. These icon sizes are super super tiny and #$&*@! as GIFs, and a vector file would go a long way to making these things actually readable.
- The main index page being uneditable means the community can't improve the new user experience. I get that the point is to avoid drive-by edits to the front page, but at the same time I don't think I've ever seen any malicious edits in all the time you've been gone (except the wet blanket that got rid of my joke image hover texts! You allergic to fun or something??), so the fear may be somewhat overblown.
- Editing history is, like, gone. I assume this is permanent, but without history a wiki is only half a wiki, and it alarms me somewhat. (And I only half want it to track down who removed my jokes.)
- Did we ever remove the post limit for new users to contribute? Again, I get the drive-by spam prevention measures, but it has also been an obstacle for users to overcome in the past, and will be so again if we get some users moved over here (that effort died in the cradle with the email broken, but with it fixed I'm foreseeing people just giving up when they finally sign in and can't actually edit the wiki for some arbitrary period of time).
-
I think your jokes went down with the edit history, unless the wiki failing was an elaborate coverup. Evidence points to Kralik. Shame that they're gone. After I stumbled across one I went scouring through the pages for images, haha.
I'm in support of a post limit, though 25 is a bit steep. What about 10? Obstacles aren't necessarily a bad thing, I'd rather have some higher quality content on a small portion of the wiki than lower quality content on all of it.
-
Teltura, responding to a few of your points, though the quote feature isn't working for me (another thing broken?):
Wiki choice: there's some history behind this.
The original lotrtcgdb.com site was born when Decipher went defunct and there was practically no place online that had all card images or text. Cobra Cards had some but then they closed as well. I didn't want the game to die so I started with an archive of card images but then added the text later so that there was more content than an image repository in case there was a possible legal issue. Creating thousands of HTML pages takes a lot of time so I wrote scripts to generate them from my card data spreadsheet, sourced from multiple places over time and obviously full of some inaccuracies as the info was at some point hand-typed. Eventually we had forums and an online client (GCCG).
At some point I decided a Wiki would be even better than static HTML pages so that players could add articles and strategy advice. I originally went with DokuWiki due to the way it handles articles as text files in the underlying filesystem. This allowed me the ability to quickly generate and import all of the card images/articles and also create backups (though currently broken) for anyone to download the information in a single file. My thought was (and still is) that if my site disappeared like the others, the game could still exist and someone else could easily remake it. Getting the old Cobra Cards working was only possible because the owner was gracious enough to provide me with a filesystem and database dump after the domain was already gone. If for some reason I stopped supporting this website, it would likely be the sort of situation where I wouldn't be able to do that myself. So basically it's a contingency plan.
I haven't looked into other Wikis for a while but my gut instinct is that upgrading DokuWiki and fixing the various issues would be a lot less time consuming than recreating the wheel, so that's likely the approach I'd take for now. Edit history and uploads being broken isn't intentional. As for the card data... I do want to keep a vetting process there as it's my goal for this site to be as accurate (as possible) and we don't have a team of editors or moderators that can review changes. I'm working on adding the Hobbit Draft cards (soonish) and 2E (if/when I get the the card text). It's possible that we could have a small team with permissions to add cards but because of how it works currently, that would require FTP access at least or possibly SSH.
Regarding the posting limit: we used to get a LOT of spam accounts... is that no longer the case? The Wiki has its own account system but I changed the code to use the TLHH login and cookies to keep things simpler. Another custom hack that I'll have to look into if I upgrade either the Wiki or the forum or both.
SVG support--not sure but maybe an upgrade will improve that? Is it an issue with embedding or the files can't be uploaded or rendered or...?
-
quote works for me.
My thought was (and still is) that if my site disappeared like the others, the game could still exist and someone else could easily remake it. Getting the old Cobra Cards working was only possible because the owner was gracious enough to provide me with a filesystem and database dump after the domain was already gone. If for some reason I stopped supporting this website, it would likely be the sort of situation where I wouldn't be able to do that myself. So basically it's a contingency plan.
So like, I get this, but if the wiki were in a format that was used by the most popular wiki software in the world, anyone would be a database dump/import away from doing the same thing. That's not that much different from an FTP link to a bunch of loose files. A bunch of loose files that require an archeologist's chisel to figure out how they work.
I haven't looked into other Wikis for a while but my gut instinct is that upgrading DokuWiki and fixing the various issues would be a lot less time consuming than recreating the wheel, so that's likely the approach I'd take for now.
You'll recall when I generated all the card articles in the first place. The export/import process would require something similar, only made easier by the existence of the wiki files as they are now. 1- prop up mediawiki. 2- write bot to read dokuwiki files and translate them into mediawiki articles 3- run bot 4- there is no step 4.
I say this having already written a mediawiki bot for the Artifact wiki (which did the same exact process, except analyzing data ripped from the game files as opposed to an existing wiki's text files). It was slightly more complicated even, since it had to be able to re-export and edit the articles in-place when game files updated with new info. This is a piece of cake in comparison.
Once a code base has gotten old enough, it can quite literally be easier to start over with a new one, particularly if that means that more than 1 person is involved with the guts of it.
Edit history and uploads being broken isn't intentional. As for the card data... I do want to keep a vetting process there as it's my goal for this site to be as accurate (as possible) and we don't have a team of editors or moderators that can review changes.
Respectfully, I disagree. Dictionary and Phallen have done wonderful sustained effort on the wiki over time, and had email notifications actually worked I would have responded as soon as I got any. There are people who could do this.
And even if we all didn't check the queue for a month, it's a *much* better state of affairs for a new, excited, energetic member to submit a bunch of edits to a queue that can then all be reviewed and utilized once someone wakes up a month later, than for that person to just be completely stonewalled, unable to contribute, they lose the spark and we lose a high potential member.
I'm working on adding the Hobbit Draft cards (soonish) and 2E (if/when I get the the card text). It's possible that we could have a small team with permissions to add cards but because of how it works currently, that would require FTP access at least or possibly SSH.
2E might be dead in the water, I would take a look before going through the trouble. I think someone said the project imploded and is only half-finished, not sure.
Regarding the posting limit: we used to get a LOT of spam accounts... is that no longer the case? The Wiki has its own account system but I changed the code to use the TLHH login and cookies to keep things simpler. Another custom hack that I'll have to look into if I upgrade either the Wiki or the forum or both.
This might actually be a case of the existing restrictions working so well that it looked like they were doing nothing, since nothing ever got through. I defer to your judgement in that case, but it's a shame when new excited members have to be turned aside.
SVG support--not sure but maybe an upgrade will improve that? Is it an issue with embedding or the files can't be uploaded or rendered or...?
It's a dokuwiki plugin that has to be installed, I believe. It was brought up a long time ago when we were setting it up, but time has only treated those dinky little gifs worse.
-
So like, I get this, but if the wiki were in a format that was used by the most popular wiki software in the world, anyone would be a database dump/import away from doing the same thing. That's not that much different from an FTP link to a bunch of loose files. A bunch of loose files that require an archeologist's chisel to figure out how they work.
I think perhaps I wasn't clear. The Wiki is currently available as a single public ZIP file that can be downloaded and easily resetup. The file hasn't been updated in a while so I need to figure out what happened, but it used to be compiled nightly. That means that anyone can download it and have an archive without needing a database dump. If I were to get, say, hit by a bus tomorrow, disappear (as per usual) and then the website closed because nobody was paying the bills, there would be nobody to make that dump. Of course a bot could scrape the pages earlier, and perhaps has already, but that was the rationale.
See: http://lotrtcgwiki.com/forums/index.php/topic,7980.0.html
I suppose I could set up a MySQL dump nightly as I already have one in place for my nightly backups. On the other hand, there's a certain simplicity with text files that can be read on every system without needing any technical know-how. I wonder if newer versions of DokuWiki would give most of the features you're looking for? The inability to edit card text is a constraint that I put in after all. It took a lot of hacks to make the wiki merge a static plus dynamic portion.
You'll recall when I generated all the card articles in the first place. The export/import process would require something similar, only made easier by the existence of the wiki files as they are now. 1- prop up mediawiki. 2- write bot to read dokuwiki files and translate them into mediawiki articles 3- run bot 4- there is no step 4.
I say this having already written a mediawiki bot for the Artifact wiki (which did the same exact process, except analyzing data ripped from the game files as opposed to an existing wiki's text files). It was slightly more complicated even, since it had to be able to re-export and edit the articles in-place when game files updated with new info. This is a piece of cake in comparison.
Once a code base has gotten old enough, it can quite literally be easier to start over with a new one, particularly if that means that more than 1 person is involved with the guts of it.
Maybe this is something you'd be interested in doing if I set up a vanilla Mediawiki install? Keep in mind though that at the moment I have some personal data on the server and can't/won't be allowing anyone else FTP or SSH access, so it would have to be possible to do from the front-end.
Respectfully, I disagree. Dictionary and Phallen have done wonderful sustained effort on the wiki over time, and had email notifications actually worked I would have responded as soon as I got any. There are people who could do this.
And even if we all didn't check the queue for a month, it's a *much* better state of affairs for a new, excited, energetic member to submit a bunch of edits to a queue that can then all be reviewed and utilized once someone wakes up a month later, than for that person to just be completely stonewalled, unable to contribute, they lose the spark and we lose a high potential member.
I agree that members with a passion should be able to contribute, but I would prefer if it was someone with a good reputation who can be trusted to keep things in order versus a random new member. One of the reasons I set up the Wiki to begin with was to allow people other than myself to make significant contributions.
If there was a smallish group of players that was willing and able to give card edits or additions a cursory look, I would have no problems with that. This website is already "running on fumes" and I haven't had the time to be actively involved other than making sure it's still up. So my long term priorities are:
1) Keep lotrtcgwiki.com with the Wiki, Forums, and GCCG client (probably unused) running indefinitely. If at some point interest drops off completely I may consider closing it, but even with low forum activity Gemp is still running strong.
2) Keep the software reasonably up-to-date and address security problems as needed. Currently the server OS is fresh and the machine has been upgraded but the forum and wiki software is very outdated.
3) Get HTTPS working. Certificates are trivial to setup but the old SMF forum software doesn't support HTTPS well so I'll need to upgrade it (#2) before moving forward there. It's likely I'd upgrade the forum first and then the Wiki afterwards.
-
I think perhaps I wasn't clear. The Wiki is currently available as a single public ZIP file that can be downloaded and easily resetup.
Ah, right. As you noted it's out of date, and has frequently gotten out of sync, I think, so I forgot about it. It's not a bad idea so long as it can be turned on and *left* on, but I think it has spent more time broken than not over the years. I always figured we'd have to crawl the wiki in the event you got hit by a bus anyway.
Maybe this is something you'd be interested in doing if I set up a vanilla Mediawiki install? Keep in mind though that at the moment I have some personal data on the server and can't/won't be allowing anyone else FTP or SSH access, so it would have to be possible to do from the front-end.
Yup, I certainly would! Bots are normal wiki users, they just have certain flags set by the admin on their account to reduce rate limits and restrictions on what articles they can modify. It won't need access to the underlying system, just the mediawiki API that everyone else can access anyway.
-
I always figured we'd have to crawl the wiki in the event you got hit by a bus anyway.
But how would you know until one day the site just disappeared? That's what happened with CC--one day it was gone with no warning--but fortunately I was able to reach the owner.
Yup, I certainly would! Bots are normal wiki users, they just have certain flags set by the admin on their account to reduce rate limits and restrictions on what articles they can modify. It won't need access to the underlying system, just the mediawiki API that everyone else can access anyway.
Great to know. I'll look into it when I get to that point. :)
-
Hey all, I've got some limited time this week and will hopefully be able to work a bit on the Wiki e.g. getting logins to work again. Message me if there's anything else that's urgent and needs my attention.