Tavern Maker

Tavern Maker is a freely available tool for all gamers and gamemaster of role playing games (RPGs). It uses manually created templates and databases to generate floorplans of taverns, which are then "populated" by fitting descriptions. Additional information, such a tavern names, rumors, menus or pick-pocket-information is generated, too.



TeamWare is mixture of Freeware and ShareWare. There is no money needed.
A small share of worktime to add to the project is all, that is asked for.

As gamemaster and player of all kinds of RPGs, I am always on the lookout for good resources on the net. I believe in the creativity of the community and don't like to spent lots of money. That's how the idea of TeamWare was born.

TeamWare means, that the programs can be shared and used freely and without any restrictions. Created content can be used & published as far as proper credit is given. Comercially use of the program, however, needs explicit permission.

Though the program is freely available, registrating is necessary. Everybody using the program frequently is obliged to "pay" by helping the project grow. This is usually done by writing a minimum of five additional descriptions for the database, but of course more help is appreciated. With every new user, the project grows and all users can benefit from this work.

To keep the project "together" and prevent splitting up into several competing versions all over the net, several affords have been made. All registered users get their own ID and new datasets can only be used with this ID until they are sent to me to be confirmed and placed on this homepage.


More than one generator...

Though TavernMaker has been created as a generator for inns and taverns, it is capable of doing much, much more!

The new version is completely open to any roleplaying system and can be adjusted to nearly everything, from a shop-gernerator to an ambush-maker, from a village-generator to a dungeon-builder.

All that is needed is a new set of databases (for textoutput) and WMF-files (for graphicoutput). The program places icons randomly accordingt to graphic-templates, calculates the distances between placed icons and uses this information together with the other settings to choose the proper text sets. With a little afford one can easily "abuse" TavernMaker and users all over the world will hopefully generate more and more datasets.

So, if you get an idea of a specific "random" generator, get in contact with the project leaders!


TavernMaker - multilingual

TavernMaker comes in English for first place.

The reason is, that the english spoken community is the biggest on the net. The program itself, however, is language independent. The language of the text output is of course ruled by the databases used, and the program's interface can be easily translated by adding a simple language-file. So if you feel for your own Italian, German, French, Spanish, Norwegian, Russian or whatever version, it can easily be done.

If there is enough interest in such a language version, enough users will be found to keep the project running. It might just need a little pioneer work first to get things running.


TavernMaker - a little history

I started the project in summer 1996. I had just finished my last year at school and was on vaccation when I got the idea of creating a program like TavernMaker. As a roleplayer and -master I had searched the internet for such a tool for quite some time without finding one fitting my needs. All programs I found only offered very "random" descriptions which lacked deepness and quality.

My idea was to create a program which only randomly selects, but uses handmade and therefore quality descriptions. Of course I wasn't interested in writing all of those myself, so I started the project as "teamware" meaning that I would offer the program for free and without restrictions, but with the request of "registering" by putting some work into the project.

Back from my vacation (-in south-Italy, btw-) I started programming at once. The first version was crude but working. It only lacked the most important thing: descriptions. To get the first version running I had to assemble about 100 descriptions and this work was mainly done by myself as I couldn't yet "publish" the project on the net. I got some help of Pete J. Mitchell and Richelieu though.

22.Februar 1997 the first working version of TavernMaker went online. And after some "promoting" things started to grow slowly. I am still graceful for all of those people who helped me getting the project out of its baby-boots. Lots of people "registered" by sending their descriptions which I then included in the database.

Even more people, however, came with "new ideas". All of them great - some of them possible to integrate into the project. So, as the version numbers grew, more and more things were added. New "races" and "tavern types", the "menu"-generator and some minor changes in the program's random algorithm itself.

A huge breakthrough was the development of WinTavern by Andrew Williams. His program was a "shell" for my DOS version, so that people had easier access to the settings under Windows environment. Unfortunately I lost contact to Andrew, so the WinTavern wasn't updated together with the main program and soon couldn't be used anymore.

Luckily I could convince my fellow student Andreas Domaingo ("Whinney") to create a new Windows-shell called WinTav, which soon became very popular and helped a lot in spreading the project.

In summer 1997 I got some big help from Jenni A.M. Merrifield who used a lot of her spare time to go over the written datasets and correct them from mistakes. Though the main version of TavernMaker was English, people submitting their descriptions came from all over the world and weren't native speakers.

Until 1998 the databases had grown up to more than 1000 entries. A "tavern name" generator had been added thanks to Siltala Timo and the "menu"-generator was now customable. The project had spread widely and several homepages started providing TavernMaker as a download. Unfortunately not all of the authors respected my wish for a "link back to the homepage" and the unavoidable happened: Different versions of the program started a lot of confusion. The positive effect of it: I met a lot of people via email....

Another big step forward was the introduction of some kind of "map"-generator in TavernMaker v4. It was crude and totally random, but at least there was some kind of map-output. However, it raised the need for a more "reasonable" map-generator and that was, when Kevin Berry ("Psyko-Kev") came into the project.

( He wasn't the first suggesting this, but he was the one who finally convinced me - maybe because of the fact that he was doing the programming? )

Anyway, there had to be fundamental changes to be made. All datasets needed some "rework". By that time (spring 1999) I was studying in Trondheim/Norway for a year and though the basic idea of "TM v5" was a good one, and Kevin spent a lot of time in the project, it was also the beginning of the project's death.

As the "new TM v5" needed different databases, I stopped updating the databases on the net. Instead I stored the new sets for later usage while step-by-step converting the old database. The work, however, took more and more time. Kevin moved around half the world from South Africa to England and I moved back from Norway to Austria. In the confusion of the move a lot of date got lost. :c(

My motivation for the project dropped dramatically, as there was no "new" version in sight and the "old" one was stuck. I simply waited for better times, hoping that Kevin would finish his work. He would, but in winter 2000 a final stroke crashed the project completely: WindowsME!

Convinced by advertising (or whatever) I installed the new software - and was, as far as posible with anything from M$, quite satisfied. The system had just one really bad side effect: All of my programs done with TurboPascal strictly refused to work at all!

Suddenly I wasn't even able to update my old databases anymore! The project had to be "frozen". Things didn't become better within year 2001. I even had to change my homepage and email, so "feedback" was dropping in frequency and I believed TavernMaker to be more are less dead.

It's due to "some" feedback and Andreas Domaingo, that TavernMaker has now reappeard from the dust like a phoenix. In summer 2001 He convinced me that "something had to be done" as there was "still a lot of interest".

After a long session of constructive brainstorming we decided to "recreate the wheel". The only thing kept from the "old" TavernMaker is its heart: the databases, but those needed to be "reworked" heavily too!

Since our first meeting in September 2001 a lot of time has passed. We spent hours, days, weeks on the project and with every new idea it became larger and larger. Whinney was doing all of the programming, while I was beta-testing, recreating the datasets, drawing the icons and templates, maintaining the homepage and so on. I'd say we splitted the working load quite 50:50.

In February 2002 we were able to present our first results. The "new" TavernMaker was still not doing half of what the "old" one did, but at least the map-creating algorithm were completed and we published it as a demo-version together with the template editing tool. Actually, we hoped that some people would link up and start creating some templates, but that didn't happen. Maybe, because it is quite a lot of work!

We turned our focus towards text-genration then, and things became bigger and bigger. First, we only wanted to "convert" the old database, but then we got the idea of adding a kind of scripting language. This made things really flexible, but it also took a lot of time to design and program all of it.

In summer 2002 I finished "redoing" the old database. At the same time I started building up a German database.

In December 2002 we finally managed to create a fully working version of TavernMaker. It still lacked some of the possiblities of the old one (e.g. menu-creation), but it outdid the old version in other respects by far. Both of us were positive to be able to "publish" the program by Christmas.

We didn't. Mainly, because we wanted to remove all bugs an still had some minor ideas, but also, because we were making up our mind on how to distribute the project. We descided that we want to make it as comfortable as possible to the users, so Whinney started programming an installing and updating tool. This, however, delayed the first "real" release of the program until Spring 2003.

Then, bejoscha got the idea of making the project more modular in order to help later updates. Whinney agreed and at once started to work. It, however, turned out that redoing the complete code was a lot of work, so the relaese date was moved towards summer... then autumn... then winter... then spring... then summer... ?

Well, and then we were finally completed: 22.09.2005 - Release of first version.

The end?

Far from! We are constantly improving, adding datasets, adding templates, adding features. However, since 2005 less progress was directly visible, as most of the work was going on behind the scenes. Large changes in our private life also forced us to cut down on the amount of time invested, and some ambitiouse projects had to be abandomed or at least put to rest due to lack of time. The calmest year so far was 2008, and it is once again Whinny's engagement which helped to rekindle the project. He was constantly working behind the scenes to get the menu-creation back into the software, following a (hopefully) sophisticated scheme we've worked out. Mid-May 2009 this add-on has reached alpha-status and looks extremely promising. What now follows is a period of testing, debugging and testing again. Once this has passed, builing up a sustainable database is the final step before releasing the add-on. For this, engagement of creative developers is needed!

Is it dead now?

2014: Almost 9 years since the 'first release' of TavernMaker now, and as you undoubtedly have noticed, a lot of dust has settled on this homepage - and also on the project. Life has moved on for both Bejoscha and Whinney, and spare time has decreased and interests changed. So, unfortunately the project is no longer alive - but neither is it really dead. It's in a zombie state (or, if you don't like horror, think of it as a Cindarella state) where bejoscha is still answering to emails and registrations, but little else is done with project....