response 50 of 83:

Feb 28 22:23 UTC 1995

I concur fully with mdw's #46. The special keystrokes are necessary
in newuser. I would suggest beyond that, that there be teh addition (gak)
of one more 4-5 line paragraph which describes/explains "what" is
coming next (in newuser) and "what" each key-choice is designed
to do.
  
At least that way, imo, the newuser can anticipate what is +about+
to happen and why it is critical that the newuser choose (and learn)
about "funny keys."

response 52 of 83:

Mar 1 10:52 UTC 1995

Is source to newuser available online?

response 54 of 83:

Mar 1 14:46 UTC 1995

Does Grex have the right to ask that source code be made
available for revisions?  Do we "own" source code?  If not
should we look into buying it?  And if source for Pico can't 
or won't be provided maybe we should look into YAPP?

I'd sure hate to see one person be able to control how Grex
looks and feels because he owns the toys.

response 56 of 83:

Mar 1 20:30 UTC 1995

I liked all the data in newuser, and learned a lot from it.  Still, I'm
probably not typical.

response 58 of 83:

Mar 1 22:42 UTC 1995

        Whatever. I just thought you might want to know the view from
this side..

response 60 of 83:

Mar 2 07:07 UTC 1995

This is a 150-line response to a mdw's long response in #46, if you
want to skip this.  I agree with virtually all of Marcus' factual
statements, and some of his ideas, but I still favor a newuser overhaul.
 
> I think Grex is particularly likely to get some of the "el bizarro"
> brand devices.  A number of people have asked about using grex from
> 3270's & MVS!
 
I gather this is a reference to suggestions not yet mentioned in this
item, but to age-old discussions where some people advocated setting
the terminal type to ANSI/VT without offering a way to change it.  If
so, I'm with you...rare though they are, I still think we should
accomodate everyone.  But the "terminal type" question could
certainly be simplified a lot, while still allowing 3270 users to
set their term type.  As carl said, make it simple for the majority.
 
> If they don't find out about control-C in newuser, they never will.
 
Agreed.  But many people never encounter the need for ^C.  It's not
needed to run the menu, picospan, party, or a ton of other programs,
it's just occasionally nice.  Sometimes people will stumble upon a
program that provides no instructions, and requires a ^C, but that's
rare.  Malicious partiers sometimes tell newbies to run such programs,
but even if newbies all understood ^C, malicious partiers could find
other ways of messing them up.  ("Hey, try this neat stty 0 command!")
(I'm not sure how common my ^C view is; I'd want to ask if other users
agree before removing the section on ^C).
 
> I would also predict that, if the section on special keys were
> removed, there would be a sharp upswing in people trying to use
> cursor left to edit the input line.
 
Sure, if you're talking about removing the section about the backspace
key.  Some people may want even that removed, but nobody has mentioned
it in this discussion so far; remmers even specifically suggested it's
a good one to keep.
 
> One of the things I notice here, is a lot of people here are jumping to
> conclusions from a very limited sample size.
 
Agreed.  But we're also basing the conclusions on experience with
computers and common sense, neither of which should be discounted just
because they're unscientific.  There is similarly no scientific basis
to conclude that newuser is ideal the way it is.
 
> People are not all the same, and just because user "a" finds it hard,
> or user "b" finds it easy, does not mean it's either easy for all, or
> hard for all.
 
True.  Restructuring newuser to provide differing levels of detail,
based on the user's choices, is one way of accomodating both "a" and "b."
 
> Despite all these obstacles, we have incontestable evidence that
> something like 70 people/day succeed in creating accounts, and we also
> have an amazingly low rate of complaints (except when it's broken - and
> then people manage to complain with no trouble at all.)
 
70 people/day run newuser and create an account, but another 50 people/day
run newuser and don't create an account.  [From Mon 3/20 00:00 to Mon 3/27
23:59), newuser was run 888 times; 9 ended in system crashes, 511 produced
new accounts (73 a day!).  This is based on the "last" and "finger" commands,
and /etc/passwd (users dacey to resonate)).]
 
There are many reasons people may have stopped running it, but some are no
doubt related to newuser's design.  My point isn't that we should to optimize
newuser for completed account creations, but that 70 new accts/day doesn't
mean people aren't having problems.
 
As for the low complaint rate, I wouldn't take that to mean people have few
complaints.  The people who bail out of newuser probably have some, but
don't typically voice them.  Also, most large companies infer that for every
complaint they receive, there are many others who agree but don't bother to
complain.  Grex isn't a big company, but I think the principle holds.
 
> I think people here are confusing the goals of making newuser easy to
> use, and making grex easy to use.  The two are not at all the same
> thing, and some of the things people have indicated elsewhere that they
> value in grex (conferencing, and, (i hope!) the diversity) - are in
> direct conflict with some of the ways people are trying to make grex and
> newuser easier to use.
 
Speaking for at least me, I recognize the difference between making
newuser easier and making Grex easier.  There are some tradeoffs, but
I think on the whole, an easier newuser would add to Grex's diversity.
 
> Let's look at another problem - that of the "." problem.  In the
> original scheme, you'll note that people go from newuser into PicoSpan.
> Newuser, PicoSpan, as well as ed, es, ex, ucb/mail, and even rcs, all
> share one common pattern for text input - a dot on a line by itself.
 
I think you missed the point (pun intended).  Newuser never explains
how to press a dot on a line by itself.  It explains how to hit
ctrl-d.  Then in its questions, it says to press ctrl-d or a dot
on a line by itself.  Picospan then says to use the dot, without
mentioning ctrl-d.  Both work, but dots seem easier to teach.  So why
not explain dots instead of ctrl-d, and why even mention ctrl-d?
(Maybe there are reasons; that's a real question, not rhetorical).
 
> Newuser is also encouraging people to open up about themselves, and
> talk about things that interest them.
 
That's a good idea, but as it is, most users (or maybe just a lot)
don't permit their .plan files publicly.  I think this is another
area where newuser could be tweaked a bit, to encourage people to
open up publicly.  One way is to tell them from the outset to just
skip the address and phone if they don't want them to be public.
I heard another idea of just saying "this *will* be public, don't
answer what you don't want to."  Still another approach is to ask
for a free-form .plan file, with suggested contents of hobbies,
birthdate, gender, city, etc., but leaving it up to the user.  There
are plenty of other options that would also get people to open up.
 
> Some of the problems people have identified elsewhere in this conference
> include: link congestion, system growth, system reliability, "problem
> users", and getting more of the people on the system into conferencing.
> By using the above systems models to analyze the changes proposed for
> newuser, we see that, virtually point by point, the changes do nothing
> to solve these problems, but on the countrary, make them worse.
 
(I may be misinterpreting your meaing, but....)  The changes proposed to
newuser are not *intended* to solve link congestion, system growth, system
reliability, and other problems.  True, they may make the problems worse,
by making Grex friendlier and more attractive to new users.  And by that
logic, we could *solve* the problems you mention by making newuser *harder*.
And it's true, we could!  But I think most people would like to see the
system easier to use, and deal with the other problems in different ways.
 
> So where do we go from here?
 
Good question. Some possibilities:
 
 * Get feedback on various ideas in agora or with the vote program,
   for example to determine how useful ctrl-s or ctrl-c are.
 * Discuss ideas like in this item, and have a staffer make a unilateral
   decision to change or not change newuser.  (Autocratic)
 * Discuss ideas like in this item, and have someone write a spec for
   approval by the board, if it's for a major overhaul.  (Democratic)
 
> Or, we could study it scientifically - have newuser randomly select a
> different mode of operation for users, then study the relative performance
> of users....
 
Interesting idea, though I think that may be overkill.  OTOH, if you
want to volunteer to conduct the study....  :)
 
> Or, perhaps, we might want to scale back our expectations a bit?
> ...even I can see things that could be improved, although I think
> my list would be a mite different than others here.
 
Personally, I hope your view and other views are all considered.

response 62 of 83:

Mar 2 18:04 UTC 1995

I think newuser is currently off-line more because the disk problem and
some other things may have damaged some of Grex's libraries that are used
to build newuser than from any intent on Marcus's part to sit on the
newuser source code.

response 64 of 83:

Mar 3 16:55 UTC 1995

  Isn't clearer easier? :)  I have no fixation on the word
easier; I'd be more interested in your opinions on specific
change suggestions.  E.g., should we stop explaining ctrl-d?
 
  Also, I agree newuser asks little, but for on-line novices,
who don't know how a control key works, it's tough.  To get
such novices involved with Grex, instead of scaring them off,
I think the info on control keys could just be skipped.  (I'm
considering "backspace" not a control key :).  I just don't
see any of the control keys as that important to using the
basic features of Grex.  Ideally, I'd say offer these novices
different info/options than are offered to more advanced users.

response 66 of 83:

Mar 3 19:34 UTC 1995

Newuser asks for an interrupt character (generally control-C) (it
interrupts the process you are currently running) and a kill character
(generally control-U) (it backspaces all the characters on the line
you are currently typing).  I don't think many new users use either
of these keys much at first.  Also, newuser makes an effort to teach 
people about control-s and control-q to stop and start text scrolling
on your screen, but these days everybody has a pager that says "press
spacebar for more or q to quit" so they don't need control-s control-q
and control-c, just spacebar and q.  If we took out all the stuff about
control characters in newuser, we could also take out the explanation of
how to type a control character (hold down the control key *and* press
the other key at the same time).  This would help shorten and simplify
newuser by a lot.

response 68 of 83:

Mar 4 05:03 UTC 1995

        I first used ^U to erase a line a couple of years ago, when I ran
"stty -a" and saw it listed, and tried it to see what it did.  I'll bet
not many people redefine it because they need it but prefer something
else.  I'll bet most people never use it at all.  Things like that, it
seems to me, can safely be left out of newuser.

response 70 of 83:

Mar 5 07:12 UTC 1995

Ive got my kill character set to delete...

Shit I just typed it and erased the above line.  I may change it one of these
days.  Anyway, the point I wanted to make was that I rarely use kill, and that 
also it does not work in all situations.  For example while hitting delete for
me here in picospan will kill an entire line, in the pico editor delete is the 
same as backspace.  No difference at all.  Maybe we don't need to tell people
about that key, However, I think that it is important to tell people about both
backspace and ctrl-c.  I use both of those all the time.  

response 72 of 83:

Mar 5 19:10 UTC 1995

Re #67 and related responses:  I'm all in favor of providing means to
help the user navigate Grex; I just think that newuser is the wrong
place to do a lot of that kind of education, as it's premature for a
lot of people.  (I realize that people's mileages vary.)  The technical
points discussed in newuser are very valuable, but they're introduced
at the wrong place.  Newuser only gives people *one shot* at it before
they're cast into the wilds of Grex, with no obvious means provided to
reconfigure things if they've made some wrong choices.

Repetition is an essential part of education, and education only
"sticks" if the person is ready for it.  It would make much more sense
to have an easily available, well-publicized command that people can
run *AFTER* they've gone through newuser and set up their account, and
that educates them about control keys and what they do, the fine points
of terminal types, etc., and allows them to configure these options.
Take all but the bare minimum of this out of newuser itself.

In fact, the "change" program is designed to be that, though I haven't
looked at it in a while.  Does it work well these days and allow people
to reconfigure all the things that newuser asks about?

Newuser should of course continue to say something about what Grex is
all about (the "declaration of principles") and list a *few* do's and
don'ts such as proposed by STeve in #0.

response 74 of 83:

Mar 6 08:40 UTC 1995

No, "clearer" and "easier" are different. Four keystrokes and
you are logged in is "easier." Forty keystrokes with supporting
reinforcement (why) is "clearer," and imo, preferable.
  
As remmers mentioned, repitition is what I'd call the key to
learning new stuff. Triple hit; tell newbies what is about
to happen and why; do it with explanation; identify what was
just done ... not too terrible for repitition, although the
third part can often be eliminated.