Fvwm History

Fvwm Timeline

History Conversation

Here is a conversation that took place around 2005 about the history of Fvwm. The conversation is between various authors of Fvwm including Dominik Vogt (DV), Charles (Chuck) Hines (CH), and Robert Nation (RN):

Dominik Vogt
DV> ((Chuck, are you still lurking?  Can you fill in the gaps before 1998?))

Chuck Hines
CH> I am still lurking, yes, but not sure how much I can fill in as my memory is
CH> getting a little fuzzy in my old age...I'll see what I can do... :)

CH> I apologize for leaving all of the cited text in here.  I normally would trim
CH> out most of it but for filling in a history it seemed appropriate to leave it
CH> in.

Dominik Vogt
DV> Robert Nation started it back in 1993.  The first time anything
DV> was heard about it in public was on the 1st of June in 1993, when
DV> Rob bundled a development version (0.5.something) with an rxvt
DV> release (a still popular terminal program).  In the next few
DV> months, fvwm became an independent package and fvwm-1.0 was
DV> released in fall, still 1993.  Originally, the "F" in fvwm stood
DV> for "feeble".  But then, Rob seems to have forgotten this at some
DV> point in time and thus the famouse FAQ question 1.1 was born.  But
DV> although there is strong evidence on the meaning of the "F" in old
DV> new group archives, we nowadays prefer the "mysterious F"
DV> interpretation ;-)

Chuck Hines
CH> I never agreed with "feeble" myself, even early on.  That was one of the
CH> reasons I came up with that list in the FAQ, to provide alternate
CH> possibilities, hopefully mostly positive.  Many of them were suggested in
CH> emails on the mailing list, if I remember correctly.  Of course, I added the
CH> explicitives to that list because they were what I most often used while
CH> trying to figure out bugs. :)

Robert Nation
RN> There were two or three reasons for starting FVWM and RXVT. First, I had
RN> a need 33 MHZ 486 laptop PC with only 4 MB of RAM, and I thought linux and X11 were
RN> way better than the windows versions of the day. X11 with TWM and xterm would run on
RN> my PC, but just barely. Second, I had a need at work to analyze spectrograms which
RN> were about 4000 x 200 pixels when displayed at full resolution (analyzing acoustic
RN> signatures for the DOD) - twm couldn't display
RN> that, although some other window managers could (they used more memory though!). Finally,
RN> I thought it would be nice to learn a few GUI software skills.

RN> I think I did rxvt first. The source code fro xterm was pretty hard to understand,
RN> so I found xvt on the net somewhere. I cleaned it up a bit and improved its vt-100
RN> compatibility in a few areas. Next, I tore apart twm to find out why it was so big, and
RN> generated a very simple, low memory, low-flexibility version of a window manager. I added
RN> the virtual desktop stuff and started using it at work too, since it could display my
RN> spectrograms very nicely.

RN> After I put the these things on the net, it was fun and educational for a while, as
RN> the programs became more capable. But as my kids got a little older and needed more
RN> attention from me), and the maintenance function got to be mostly integrating patches
RN> from various people (and the patches had little or no utility to me), the fun went
RN> away, and Chuck Hines stepped up to take over.

Dominik Vogt
DV> I know litte about what happened between 1993 and 1996.  Rob
DV> stopped maintaining fvwm and Charles Hines took over the project
DV> for several years(?).  When Chuck resigned, Brady Montz became
DV> the new maintainer for a couple of months, I believe in late 1997
DV> or early 1998.  Anyway, it took almost 8 month between the 2.0.45
DV> release (22nd of January, 1997) and 2.0.46 (20th of August, 1997).
DV> I can only guess why, but probably this was a foresign of Chuck's
DV> approaching retirement.  At this time I had been using fvwm-2.0.x
DV> for about two years at home and wanted to implement some of the
DV> features I liked in CDE to fvwm.  In December 1997 I sent Brady my
DV> patches and never got an answer.  I tried again half a year later,
DV> and contacted the mailing list, but fvwm development was as dead
DV> as it could be.

Chuck Hines
CH> I believe that I maintained fvwm from Aug 95 through May of 98.

CH> I had been using it actively from an early point (probably right around the
CH> 1.0 version, but I can't be sure, I have some saved emails about fvwm from
CH> early in 1994 and I know I had been using it for a while before then).  I was
CH> working at IBM and the machines we had at the time while fairly powerful were
CH> still brought to their virtual knees by mwm, so I believe I searched USENET
CH> (comp.windows.x.apps) for a good replacement.  Found fvwm and never looked
CH> back... :)

CH> I kept a close eye on fvwm after discovering it (I believe there was a mailing
CH> list at "fvwm@shrug.org", and stuff on comp.windows.x.apps of course).  I
CH> contributed a couple of minor patches to fvwm in those earlier days, like to
CH> get it to compile nicely under AIX and some stuff with colormap behavior (I'm
CH> so happy to not have an 8bit display any more).  By the way, if anyone has (or
CH> can track down) any archives of that earlier mailing list, it'd probably be
CH> pretty cool to resurrect them...
CH>
CH> Then one day Rob had sent out a message (on the mailing list, I believe) that
CH> stated something to the effect that he didn't want to work on fvwm any more,
CH> and was looking for someone else to maintain it.  I wanted to offer to take it
CH> over (as an X11 learning experience - I had dealt with X at the toolkits level
CH> but was curious about the more low level stuff) but didn't feel I had the time
CH> to properly devote to it so I didn't say anything at first.
CH>
CH> Then after about 2 weeks of watching people say "I'd love to, but I can't
CH> program" I figured I'd better make that offer after all, qualifying it with
CH> something like "I don't know if I have time to do this right, but I'm willing
CH> to make an attempt".  Rob remembered my previous contributions and figured I'd
CH> be able to make a go at it, so he turned it over to me.  That was version
CH> "pre-2.0-patchlevel-33" if I remember correctly.
CH>
CH> Then I worked on it sporadically over the next couple of years.  At first I
CH> wasn't too bad about getting releases out, but the length of time between each
CH> one got longer as "real life" invaded.  Another part of the problem was
CH> because I did use it as a learning experience, I would often rewrite patches
CH> that were submitted to me so I really understood what was happening and with
CH> an eye towards making things "easier" in the future (which I'm sure pissed off
CH> a couple of people, but at the time I thought I had good reasons for doing
CH> things that way).
CH>
CH> Finially I had to admit that I just didn't have the time to properly devote to
CH> it any more.  So it May of 1998 (I believe) I sent some email privately to
CH> Brady Montz, as I felt that the stuff he contributed showed him to have the
CH> most promise for taking it over at the time, and he (reluctantly?) agreed to
CH> give it a shot.
CH>
CH> And you know the rest of the story from there...  I wish I could have done
CH> (and learned) more, but I'm happy to have done my part.

Dominik Vogt
DV> Then - I believe it was in September or October (I have to look
DV> this up in the mailing list archive) - Brady resigned and
DV> everybody on the mailing list thought this would finally be the
DV> end of fvwm development.  Seeing this and not willing to give up
DV> fvwm this easily, I took over the job as fvwm maintainer for the
DV> moment.  After a lot of discussion we agreed that we should try to
DV> make a first stable release in the 2.x series as soon as possible.
DV> With a great team effort we were able to resolve the most pressing
DV> issues and managed to get the stable 2.2.0 release out the door
DV> in February 1999.  (I want to thank all the people who helped to
DV> make it possible back then).

DV> My memory of the sequence of events is a bit foggy.  Some time in
DV> late 1998 or early 1999 we decided that having a single maintainer
DV> as the master over the code wasn't such a hot idea.  In the past,
DV> people had been eager to work on fvwm but had been hindered by the
DV> maintainer, or more precisely:  by the absence of a maintainer.
DV> As a result, almost no work had been done in fvwm for over one and
DV> a half years.  So I gave up my role as the maintainer and the
DV> responsibilities were since taken over by the people on the
DV> fvwm-workers list.  Unfortunatly I'm still stuck with most of the
DV> project organizing work.

Chuck Hines
CH> I have to admit that at the time I didn't think the "rule by committee" stuff
CH> was going to work with fvwm, but all in all it seems to have worked out pretty
CH> well.  I think that a lot of that has to do with the fact that you were there
CH> to steer it's course.  Having some other pretty sharp individuals contributing
CH> helps too, of course. :)

Dominik Vogt
DV> Anyway, since that day in October 1998 when a complete X newbie
DV> couldn't keep his mouth shut fvwm development is as active as it
DV> will ever be.  Some of the people that helped to make 2.2 possible
DV> left and some became less active, but we still have an excellent
DV> team here (with a weakness in writing documentation).  Although
DV> I'd rather not single out anyone, I have to mention Olivier
DV> Chapuis and Mikhael Goikhman who are doing an excellent job in
DV> providing some framework (configure, fvwm-themes,
DV> internationalization, not to mention all the nifty features in the
DV> fvwm core they have written).  Although at times it might appear
DV> as if I am the one that keeps fvwm alive, fvwm wouldn't be half as
DV> good without the help of the many people on the mailing lists, be
DV> it by writing bug reports, complaining about missing features,
DV> answering questions of other users or simply encouraging us to
DV> continue our work.

DV> What about the future?  Well, the work for the next stable series
DV> (2.6.x) is proceeding very well.  Fvwm will go into feature freeze
DV> again near the end of the year so that 2.6 is ready before fvwm's
DV> tenth birthday on 1st of June, 2003.  I have vague plans for a
DV> big event on that day to remind people that fvwm is still there
DV> and that it can easily compete with any other window manager.
DV> After that there are plans for a version 3.0 that would change a
DV> lot of the syntax and introduce fantastic new features, but that's
DV> too far from now.

Farewell
CH> Ah the future...scary and exciting... :)
>> Perhaps it'd be worthy of a page on the FVWM website too?
DV> That is definitely going to be done for fvwm's birthday.
CH> Cool.

CH> Hope my feedback helped fill in some more of the blanks.

CH> Later,
CH> Chuck