Return-Path: X-Sender: fms@mailhost.cam-orl.co.uk Date: Sun, 07 Sep 1997 17:30:28 +0100 To: su1244@eclipse.co.uk, Frank Stajano From: Frank Stajano Subject: Re: DHT and Heretic Cc: Jeff Makaiwi At 19:08 01/09/97 +0100, Robert Reevy wrote: >Hi Frank! > >Sorry for the delay (been on holiday). > >I have been thinking. > >In the old days (I mean mid-late 1996) I really enjoyed following the >bases (for DOOM Incubus and later DOOM2 Incubus). It was really fun >because NM is so interesting. I think that for Heretic there are >advantages and disadvantages in having bases: > >ADVANTAGES: > >It would be fun for people to watch and see what happens, it would be >like the old system all over again, I really enjoyed it). > >It would make building the level sets easier and quicker. > >DISADVANTAGES: > >It may well be hell to maintain. > >-------------------- > >So here is my proposal: > >I start the system for Heretic Master, Pacifist, Tyson and Incubus, >based on the rules I sent you. I type up some official rules and >constraints. > >People e-mail ME for the challenges, I issue them somehow (I may need to >ask somebody else to write a program that reads my bases file in and >assigns levels from them - not easy with Visual Basic). > >The completed exams are sent to me, I do the latest log + hall of fame >and send it to Jeff with the exams, which are made available on >cdrom.com. > >No DONE forms are necessary this time. > >I recommend we have a separate directory for the HHT exams at this >stage. Then when the bases get full, we get up teams to design level >sets. Until then I will do as much of the maintenance as I can. > > >Do you think this is a good idea? Sadly no -- but let me elaborate. It is very pleasant and flattering for me to see that there are people sufficiently interested in DHT to propose extensions such as you're doing. However I want to keep enough control over it that I can make it be what I meant it to be, i.e. being in the spirit of the original rules etc etc (call this "benevolent dictatorship" if you like). Having run this thing for more than 3 years now, I have a pretty good understanding of the issues involved and of the amount of work that goes on to maintain such a structure. For this reason I will not allow any addition that involves extra human maintenance to be run. I can see that you are very keen and willing to dedicate much of your time to this (and accept my sincere thank you for this!) but I don't want to force you (or anyone else) to a committment such as (say) to issue challenges by email to whoever requests it at least every week for the next 5 years. You may have lots of spare time RIGHT NOW but may find that you have much less when you fall in love with Patsy Kensit, get a job at Id Software or are accepted for a PhD in Egyptology at the University of Cambridge or (add your favourite unexpected pastime here), and I'd be in trouble to find someone as dedicated as you to replace you, all the while having to handle all the requests myself. This is why I do NOT think your proposal is acceptable for the DHT. I am trying to move towards a fully automated system and won't like any proposals that "undo" what I'm trying to do. I only write new bits of DHT software every now and then, but if you look carefully you see that they are all part of a grand design that was laid out more than a year ago (when I decided that I really couldn't stand the manual maintenance any more). This doesn't mean, though, that I won't be open to suggestions; contrariwise, I think in fact that you have some good points in your proposal and it is worth saving THOSE if we can weave them into an automated solution. You say you liked the base concept and the progression that filled the harder slots. What do you think of the following counter-proposal? Suppose the hall of fame (which, when I write my robot, will be updated every night) has, for each title, an html table of all the maps saying, for each map, its rating (Easiest to Hardest) and a possibly empty ordered list of all the players who did that map for their exam for that title (with links to the downloadable exam of course). Then people could see which maps were still empty (sort of "out of base") and try to add to them (possibly with the "free" if they don't get it in their exam) so as to get their name in first. Say also that the first one for each map would get his name in gold, or something like that. How do you like that? It has most of the benefits of the base system, but is compatible with my automation targets. I am of course open to modify this in any way, provided that you can persuade me it's an improvement! I am open to even much more complicated (and more base-like) changes to this, even if they lead to a relatively more complicated back-end program (within reason), as long as: 1) they are sufficiently simple and unambiguous for the poor USER and of course 2) they don't require manual interventions from the maintainers. I am thinking of things like submission of the exam to a CGI program that timestamps it to say when it got in, etc etc. Frank (filologo disneyano) http://www.orl.co.uk/~fms/