The GenWeb list grew rapidly and carried over 120 messages in October,
1994. This is Part II of the October 1994 archive file.



          Thursday, October 13, 1994 12:25:35 AM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb
In article  Chris Garrigues  writes:
>The remaining concern is that I still don't want other people adding 
>individuals to my database, although adding pointers to their own individuals 
>is fine with me.  This is because I don't particularly want anybody other than 
>myself to decide to double the size of my database.  On the other hand, if 
>they add pointers, that is in the spirit of the web.

CHRIS:  I plan to set up computer systems sponsored by family organizations 
like the Austins which will start out with perhaps several thousand names, and 
hopefully grow to 2 to 10 million linked names with the help of thousands of 
people who will link in to these nodes on GenWEB.  We'll plan to have a lot of 
unused disk capacity when a node goes on the air to provide room to collect 
all possible information on the particular families.  I expect many families 
will follow suit.  What better way to get national family organizations 
working together. 

>> also like to see E-Mail between researchers and cousins relating to an 
>> individual kept with that individual's file for posterity, to be kept until 
>> primary sources are located (if ever) and put the conflicts to rest.
>Part of this I would imagine would be to define an email alias for every 
>individual in the genweb database which would send to everybody who has an 
>interest in that individual.

>At sites where the genweb manager has control over the mail aliases database 
>as well, this should be fairly easy (it's a SMOP). However, I'm not sure 
>that'll be true at all sites.

>I could see a mail-to URL in each record which would point at the appropriate 
>alias.  Using forms, there's probably also be a way to automatically add 
>yourself as an interested party.  Doing this would add you to the alias list 
>for that individual and send automatic email to those who are already on the 
>list letting them know that they have yet another cousin.  I suppose it would 
>propigate to that individuals ancestors as well.

CHRIS:  Thanks for your specific suggestions in this area of setting up for 
email lists associated with an individual.  This will be a natural and most 
effective way for cousins to find eachother and start working together on 
documenting and linking-in common ancestors.

How do we create these mail lists and email data bases using LifeLines to 
store other individual data ?   perhaps TOM WETMORE can reflect on 
this when he returns Friday from New Jersey.  

BIRGER WATHNE:  do you have thoughts on how the individual Email lists and 
storage could be implemented?  

Regards,  Bill Minnick



          Thursday, October 13, 1994 12:46:25 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb

With access to a UNIX node, it would be easy to set up a listserv,
and have links in a person's HTML page to
- get a form for posting to the mail-list
- get a form for subscribing to the mail-list (And perhaps unsubscribe as well)

These forms would just be something that lets you put down your
- Name
- Email address
- Affiliation with the person (if any)
- Text
and mail this info to the appropriate mail-address.
Just as you subscribe/unsubscribe to genweb by sending mail messages to
the automatic listserv.



Birger



          Thursday, October 13, 1994 1:52:34 AM
          GenWeb Item
  From:           Gene Stark,starkhome!gene@sbstark.cs.sunysb.edu,Internet
  Subject:        Re: LifeLines -> WWW gateway
  To:             GenWeb
Bill Minnick writes:
>In the interest of simplicity, let's proceed with the html page 
>format you have now. It appears readable, useful and quite adequate to get 
>started.  I will be interested in varying the format later when/if it becomes 
>obvious that user interface will be easier/better with specific page format  
>changes. 

>When you are finished with your formating, what files will I ftp back from 
>you?  Will your files be targeted to James Jones' 486/Unix system?  How will 
>your effort differ if we wanted to put this data base on a DOS-based system?

I have made the processed "wwwra" database you sent me available for
anonymous FTP at the following URL:

	ftp://starkhome.sunysb.edu/pub/wwwra.tar.gz

The file is 993236 bytes, and it is a gzipped tar file, made up of
several thousand HTML files.  Unfortunately, this is on my home system,
and the network link is a 14.4K dialup connection, so it will take
you about 15 minutes to transfer this file.  I tried to put it on
the CS department's anonymous FTP area, but it is full.

Uncompress the file by using "gunzip", and then extract using "tar".
If you don't have tar, let me know and I'll think of some other way to
package it.  The material was prepared on Unix, so the individual files
don't have carriage returns (^M) at the end of each line, just newlines (^J).
This doesn't usually cause any problems for DOS-based programs, but if it
does, let me know.

See how you like the format, etc., and see if the notes information has
been completely included.  I will incorporate any changes into a new version
of my processing program.

							- Gene Stark


          Thursday, October 13, 1994 2:41:46 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb
Kurt:

>I really wish that genserve was up and running for two reasons (yes, this
>is posted with the right subject and to the right list):
>
>1. I am anxious to try it out as yet another way to disseminate and gather
>   genealogical information on line.
>
>2. I think that a lot of these questions about how genweb should be structured
>   could be answered with ``well, the genserve way is ...''
>
>3. I don't really understand why genweb needs to be anything except an
>   alternative interface to genserve (of course, html-generating reports
>   would be used rather than ascii-text generating ones, but hey :^) with
>   genserve databases located at multiple sites (since this is the
>   web-way).


I agree - in fact some months ago I tried persuading Cliff Manis that a
common WWW interface to GenServ and the Roots Surnames list, which shileded
users from the fact that two separate databases were involved, was an
intriguing possibility. (I've probably got my old messages on this
somewhere still.)

Cheers

Brian

Dept. of Computing Science, University of Newcastle, Newcastle upon Tyne,
NE1 7RU, UK
EMAIL = Brian.Randell@newcastle.ac.uk   PHONE = +44 91 222 7923
FAX = +44 91 222 8232



          Thursday, October 13, 1994 8:53:47 AM
          GenWeb Item
  From:           Floyd Nordin,fnordin@ix.netcom.com,Internet
  Subject:        Ancestral File Problems
  To:             GenWeb
I have been a Lurker up to now.  I am the president of the Silicon Valley
PAF Users Group. We have a BIG interest in GenWEB. We also have 1600
members in our group. We have been helping our members to improve their
computer skills w.r.t genealogy for five years now.  I was a director of
an LDS Family History Library for 3 years. I have had a consulting business
for over 20 years doing harware designs in the early years, but for the
past 10 years have done all kinds of software projects.

I would like to share some thoughts and experiences concerning the
"Ancestral File" owned by the LDS Church. Most of you are probably aware
that it contains over 13 million lineage-linked names and is distributed
to the public via 4 CD ROM's. It is compressed data and must be accessed
with the companion "Family Search" software. It is only available in one
of the 2000 or more LDS branch Family History Librarys (although a pilot
program is now being conducted with 1000 homes to see if Family Search is
good enough to be used in the home). It is also available in several other
public libraries like the Library of Congress. As you probably know the
Ancestral File is just one part of the Family Search package. There are
also 38 other CD's with other non-lineage-linked genealogy data on them.

There are several problems with the Ancestral File system that I am
aware of that should & could be avoided by GenWEB.

First, It is owned and maintained by the LDS Church, not the authors
of the data. The problem is they cannot merge all of the duplicate
individuals that come in thru data submissions from authors. They have
been trying to solve this problem for years. I don't think it can be,
at that level.

Second, original source citations in the notes are not included in the
distribution so it is impossible to judge to "correctness" of the data.

Third, another major problem occurs when someone downloads data from the
Ancestral File and then sends in "correction" data. If they use the PAF
program, it limits all name pieces to 16 characters. Thus when they
import Ancestral File data thru GEDCOM the names that are longer than
16 characters are truncated. So far - no damage to the Ancestral File,
however if that individual corrects some of the data and sends it in
to Ancestral File and it is incorporated, that original name is forever
destroyed by the truncated version which over-writes it.!!

We of the Silicon Valley PAF Users Group believe that the Ancestral
File is very valuable as a starting point for researchers. It does give
the names and addresses of the data authors of downloaded data so they
may be contacted for original data. However genealogy searchers would
be better served if they could browse and download data with source
citations, pictures & sound, as is being proposed by GenWEB, from the
individual sources of that data - direct. Any data found would be the
"latest & greatest" from the best authority on that data - the author.
Mini-databases would be easier to evaluate by browsing patrons as being
"good" data by the quality of the source citations etc. This would be
added pressure on authors to be sure their data is "ready" for public
review rather than seeing how many names they can throw into the pot.

I see GenWEB as being a great service to genealogy data authors
(researchers) that usually don't know much about computer technology.
It can help them get their data converted from their world into GEDCOM
then into the form that can be accessed on the Internet by the masses.
(Most of us will never be researchers. My wife is one, and it takes
special skills and dedication.)

I Hope the GenWEB project is successful. I will help any way I can.

			Cheers, Floyd Nordin, fnordin@ix.netcom.com



          Thursday, October 13, 1994 9:27:06 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb
In message , Bill Minnick typed:
> CHRIS:  Thanks for your specific suggestions in this area of setting up for 
> email lists associated with an individual.  This will be a natural and most 
> effective way for cousins to find eachother and start working together on 
> documenting and linking-in common ancestors.

I'm glad you agree with this.  It's such an interesting idea and is such a 
different model from what we've all been doing with email and newsgroups.  I 
suppose the people on soc.roots who object to splits on ethnic or national 
lines would really dislike this idea.  ;-)

My only concern is that we'd be creating SO MANY mailing lists.  If you've got 
a system with several family databases each with several thousand individuals, 
this means having many thousands of mailing lists on that system; most of 
which would be getting zero traffic.  I hope existing software can handle this.

A question for Birger:

I've never set up a listserv.  How many mailing lists can it support on a 
given node?  If I have 10,000 mailing lists, but only 4 or 5 are in regular 
use, will I have to worry about slow searches of the list of lists or has this 
been at all optimised?  (My hope is that the mailing lists are in a hash table 
giving a near constant time access; my fear is that it's an unsorted list or 
array given a linear time access.  With this many lists even a sorted list or 
a binary tree could be a problem with a log time access.)

(As some of you have probably noticed, I'm very concerned about performance 
issues.  I'd hate to come up with a design which works fine with a few dozen 
databases or a few tens of thousands of ancestor records, but which doesn't 
scale to thousands of databases or hundreds of millions of ancestor records.  
I really think that in a fairly short time, genweb could be *much* larger than 
the LDS ancestral file.  Maybe in a few years, the LDS will be making an 
effort to join us. ;-)

Chris


Chris Garrigues                              (MIME capable) cwg@mcc.com
Microelectronics and Computer Technology Corporation    +1 512 338 3328
3500 West Balcones Center	                    Fax +1 512 338 3838
Austin, TX  78759-5398          USA




          Thursday, October 13, 1994 9:31:59 AM
          GenWeb Item
  From:           baud@research.att.com,Internet
  Subject:        PAF vs. GEDCOM (was Ancestral File Problems)
  To:             GenWeb
It is very exciting to see Floyd here and excited about genweb. The SVPAFUG
has been a driving force behind many developments, including attempts to 
standardize source information formats in PAF, and organizes lots of folks 
with common interests. I used to be a PAF user, and looked to groups such 
as SVPAFUG to give me guidance. As an example, Floyd's comments about
the AF are quite correct.

However, I am a PAF user no longer, and I advise caution in looking to
guidelines derived from PAF as useful in the wider world. Floyd himself
points out the AF problem, but in the update system the fault lies not
with the AF, but with PAF itself, the input medium. The medium we must
consider is not PAF, and not SVPAFUG-GEDCOM (they use PAF+NOTEs to
do everything you'd like to do with other record types), but GEDCOM 5.3
(full use of legal record types).

I am now a confirmed lifelines user, and, biased as such, I feel that
the appropriate configuration of genweb is as (simply) another interface
to genserve (still wishing it was up and running :-). Consider this:

	The www has only recently become graphical (In fact, line
	browsers still exist for use on dumb terminals).

The point of this is that genweb should be thought of as simply a graphical 
interface sitting on top of some database engineering, just like everything 
else on the web. Genweb should not be thought of monolithically. Consider
the following ascii drawing, which I hope gets my point across:


	mosaic -> query -> database engine -> response -> mosaic 
	(user)    (http)     (lifelines)       (html)     (user)
				  |
				  v
			     database(s)
				  ^
				  |
	                     (lifelines)
				 ^ |
				 | |
	mosaic -> submission ----+ +----> withdrawal -> mosaic
	(file)    (ftp:gedcom)            (ftp:gedcom)  (file)

This system currently exists with genserv (if it was up), just with

	command line/vi instead of   mosaic	for	user interface
	mail/ascii	    "	   mosaic/http	 "	query
	mail/ascii	    "	   mosaic/html	 "	response
	mail/file	    "	   mosaic/ftp	 "	submission
	mail/file	    "	   mosaic/ftp	 "	withdrawal

I would contend that a mosaic interface to the existing genserv system
would be the best basis for the genweb system, since genserv is currently
the best (designed anyway, although not running, while RSL and other things
are running and so might be considered better :-) system in existence.

If you could design some non-graphical system better than genserv, then
I would consider making this system the basis of genweb, but right now
I can't since no such system exists.

I'm off to Germany to see if I can find some relatives!
kurt :-)


          Thursday, October 13, 1994 10:06:34 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Modification of other's databases (was Re: lifelines (Why Go
Public
  To:             GenWeb
In message , Bill Minnick typed:
>     One can start from the premise that "my data base is perfect; others may 
> look at it, but do no touch!"  My experience says that that attitude misses 
> the great opportunity that GenWEB presents.  I don't care if you think you 
> have everything documented about your parents, grandparents, and so on, 
> someone you don't know out there has some additional info on people in "your"
+  
> data base, and I believe in making it easy for anyone to attach that 
> information to the individual record.  I agree that there will have to be 
> limits in the size of data files appended to the individual record; perhaps a
+  
> simple reference could be made to a new source of a lengthy biography 
> or extensive photos or movies that exist elsewhere regarding the individual. 
+  
>    Certainly we'll request a source citation be entered for any change in 
> vital information on a person.  If current vital info is from primary sources
+  
> (birth, marriage, death certificates), then we would like to have a provision
+  
> to lock those elements of the record.  but I say that any record supported 
> only be secondary or tertiary sources should be left open for correction when
+  
> better information is found (by anyone). 

I think this topic needs to be well discussed by everyone on the list ,not 
just the 3 or 4 of us who have already been blathering.  This includes the 
lurkers who don't have the technical knowledge to contribute to an 
implementation, but do have some idea of what they think GENWEB would have to 
be in order to be useful to them.

What follows is a stream-of-conciousness ramble on various part of this topic.

As I've said elsewhere, my opinion is that anybody should be able to add links 
to other databases, but they shouldn't be able to change or remove or add 
larger pieces of data.  When these links are added, a proper audit trail needs 
to be constructed.

In addition, I think that a change to the database proper should be only made 
by the owner of the database.

An interesting side question is how much we can or should control the owner of 
a database's ability to remove links that someone else added to their database.

Suppose I have a database which has some information in it which is just plain 
wrong, but I'm too stubborn to admit it.  Someone else adds a link from my 
wrong data pointing to her corrected data.  I don't like this pointer because 
I disagree with the conclusions that it points to.  Should I be able to remove 
the link from my database?

Now, I wouldn't be this stubborn, but someone else might be.  (My grandfather 
did *not* have a child out of wedlock!  How dare she suggest that he did!)

It's possible that this problem may be self policing.  If a database owner 
removes links from his database, someone else may clone the good parts of the 
database, and people will cease to link into that person's database.  (of 
course, that stubborn person can (and probably will) create his or her own 
links back into GENWEB proper . . .)

Also, can we assume that everybody poking at the databases wil be doing so for 
proper reasons?  I can imagine some jerk adding random links to a database for 
no more reason than the fact that he can.  I hope this doesn't happen, but 
people have done worse things to databases and computer systems on the 
Internet already.

Is the solution to this problem the disallowal person A be from adding links 
between person B's database and person C's database?  Maybe you have to be the 
owner of the database that any link you create points to.

In it's most basic form, what we're proposing is a total anarchy where anybody 
can provide their own data and anybody else can modify this data.  Such an 
anarchy might work.  It also might not.

I think we need some enforced structure, but we need to leave as much 
flexibility as possible in the appropriate parts of genweb.

Opinions?

Chris


          Thursday, October 13, 1994 10:11:19 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Using WWW as an Interfacing Tool
  To:             GenWeb
From the responses I've had to my note in which I mentioned earlier
discussions I'd been involved in re a WWW interface to Genserv, I thought
it worth resurrecting some of the messages I exchanged with Cliff Manis and
John Chandler on this and related topics.

The discussion, which took place last June, in fact arose out of my
concerns regarding the incoherent set of genealogical newsgroups/mailing
lists and information services that a would-be user was (is!) faced with. I
suggested that, for example, the Roots Surname List could be converted into
GEDCOM and added to GenServ. John Chandler did some interesting experiments
on converting RSL to GEDCOM for this purpose. Cliff very understandably was
concerned that simply adding such data into Genserv this would "pollute" it
with a lot of in some sense "low quality" information.

This led to my suggesting that it would be, at least in principle, feasible
to provide a front end, at a different site even, which would make an
RSL-derived service and GenServ look to a user like a single service. (I
also suggested doing the same for TMS.) We then went on to discuss briefly
the notion of using a WWW server as the basis for such a front end. What
follows is an edited set of extracts from some of the messages I exchanged
with Cliff and  John.

Basically, what the discussion was all about was the issue of how could one
separate out issues of interfacing from issues related to information
storage and retrieval. (This of course is simply following the standard
computing science "philosophy" of trying at all times to treat separable
logical concerns separately - on grounds of avoiding unnecessary system
complexity and unnecessary implementation effort.)

I have not been following the Genweb discussions for long or in detail - so
I apologise if some of these points have already been debated ad nauseum,
but I have the impression that at the moment some at least of the
discussion is treating issues of data and interfacing in a more deeply
inter-twingled fashion than is really necessary.

Cheers

Brian


          Friday, October 14, 1994 8:10:15 AM
          GenWeb Item
  From:           Amelia Painter,apainter@coyote.csusm.edu,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go Public With Dat
  To:             GenWeb
In reference to Chris's above noted message (one of the few I've 
understood enough to make a comment on)....as a family history researcher 
I want to know WHERE data comes from and HOW I can verify it.  I was sent 
down many an incorrect path in the early days of my search for long-death 
relatives!  Heresay is a nice starting point, but I want to know where I 
can write or go to obtain an actual copy of any document used in making a 
connection to my family line.

Also in reference to an earlier note regarding Family Associations.  I am 
the Editor of the Chapman Family Association's Newsletter and on the Board of 
Directors of this national organization.

Amelia Chapman Painter
apainter@san_marcos.csusm.edu
PO Box 154
San Luis Rey, CA 92068


          Friday, October 14, 1994 8:21:41 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Re: Changing Data
  To:             GenWeb
Bill:

>KURT:  Then would someone among us get a clone of GENSERV running and let us
>evaluate it first hand.  Is the software available for several of us to set up
>pilot data bases for evaluation of the concept?   If not what will it take to
>get it? Is Cliff Manis the developer/ owner,   and is he willing to pass the
>software along? -- -- Bill Minnick, Cupertino, CA

The present situation re Genserv is described in the attached message from
Jeri Steele who is now in charge of it while Cliff Manis is off in Korea. I
know she would welcome some help with some of the UNIX- (as opposed to
Genserv-) related problems she has been struggling with. (I rather doubt
that Cliff would be very keen on distributing any copies of his system.)

Cheers

Brian


====

Date:    Thu, 18 Aug 1994 01:39:46 -0500
From:    Jeri Steele 
Subject: GENSERV/GEDCOM

Joan and others that are wondering about GENSERV.

GEDCOM files can be submitted via snail mail to:
        Jeri Steele
        P.O.Box 110480
        Carrollton, TX 75011-0480

Yes, a GEDCOM file from Brother's Keeper will work.

[For those that don't know, GENSERV is a GEDCOM matching system. You submit
a GEDCOM file, to get an account.  Then you may send queries to the matching
system to get family group sheets, lists of surnames, etc.  GEDCOM files
remain your property, I do NOT release them to anyone in GEDCOM form.
GENSERV was conceived by Cliff Manis and programmed by his friend Ron McDowell.
I have been entrusted with this system until Cliff returns from Korea.
If you have a GEDCOM file on the system you will receive email from
me verifying your participation when I have the system back up.]

At this time please do NOT send any GEDCOM files via email. I am working
as best I can to get the mail up between gentech.org and metronet.com.
When that happens I will let people know that the GENSERV is back up
and I will accept GEDCOM files via email at that time.

Sorry, its taking so long, but when I agreed to move this system from cliff's
machine, I was  not counting on having a two-level fusion of neck vetrebrae
in June.  I made NGS, returned home and went into the hospital the following
Thursday.  I will not be allow to return to full time work for another
2 weeks, so I'm limited by the number of hours I can sit at the keyboard.

Thanks for your patience.

            Jeri Steele




Dept. of Computing Science, University of Newcastle, Newcastle upon Tyne,
NE1 7RU, UK
EMAIL = Brian.Randell@newcastle.ac.uk   PHONE = +44 91 222 7923
FAX = +44 91 222 8232


          Friday, October 14, 1994 8:24:50 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go Public 	 With
  To:             GenWeb
Hi Chris:

>In it's most basic form, what we're proposing is a total anarchy where anybody
>can provide their own data and anybody else can modify this data.  Such an
>anarchy might work.  It also might not.

Were I betting person, I'd take a large bet that it wouldn't! :-)

I'd prefer to look at the problem by taking the world of
books/articles/etc. (hereafter just termed books) as an analogy. Books get
written and placed in libraries, but it is not the done thing to go round
adding annotations to other people's books. Instead what one has are
various tools and techniques for (i) making references in one book to
information held in another book, and (ii) for finding out what references
have been made to a given book. I'm not claiming that the existing tools
and techniques are perfect, but they operate on a heroic worldwide scale. I
suggest that pursuit of this analogy will help you think through just what
problem you should be trying to solve and how best to solve it (the best
solutions are often those that one has managed to steal from someone else
:-)

Cheers

Brian


          Friday, October 14, 1994 8:27:20 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Re: PAF vs. GEDCOM (was Ancestral File Problems)
  To:             GenWeb
Re: Kurt Baudendistel's message of Thurs 13 Oct:

>The point of this is that genweb should be thought of as simply a graphical
>interface sitting on top of some database engineering, just like everything
>else on the web. Genweb should not be thought of monolithically.
.....
>I would contend that a mosaic interface to the existing genserv system
>would be the best basis for the genweb system, since genserv is currently
>the best (designed anyway, although not running, while RSL and other things
>are running and so might be considered better :-) system in existence.
>
>If you could design some non-graphical system better than genserv, then
>I would consider making this system the basis of genweb, but right now
>I can't since no such system exists.

Kurt has made, rather better than I did (and graphically, even :-) some of
the points I was trying to make in my message of yesterday "Using WWW as an
Interfacing Tool".

I would just add that Genweb should, if possible, be made to act as a front
end not just to one data repository or type of repository but rather to as
many worthwhile repositories of genealogical data as possible. The "owners"
of these repositories and the data that is held in them are by no means
likely to want to merge their data, or convert to using a common system,
but a good interface can shield users from the differences between the
systems, and ideally even the fact that multiple systems exist. (Indeed I'd
suggest that one of the very important resons for the success of WWW was
the fact that the people at CERN who designed it thought very deeply about
how to maximise the utilisation of existing systems, and minimize the
amount of code that has to be implemented to create WWW.)

Cheers

Brian



          Friday, October 14, 1994 8:42:44 AM
          GenWeb Item
  From:           Roger Prince,trpri@chevron.com,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go
  To:             GenWeb
Chris pointed out that letting everyone modify everyone else's data would be 
anarchy, and Brian used the library analogy to make the point that you don't 
write in other people's books, you write your own book and reference them.  I 
had been thinking of the library analogy already from the security point of 
view.  Except for some local school boards, most of us don't think librarians 
should be in the business of censoring books, and in particular we don't 
think they should be expected to nor do we want them to be arbiters of the 
truth.  Besides being an impossible task, it would be pretty strange to have 
librarians annotating or blacking out passages in the books in their care, 
even stuff they were sure was absolutely wrong.  Likewise the solution with 
genealogy data isn't to insist on one right answer in the database--an 
unachievable goal anyway--but to provide a robust, multipathed, 
multisolutioned network (sounds like internet!) with as much documentation on 
sources as possible and with powerful user-friendly indexing, searching and 
retrieval functions.  

Bill Minnick asked what sort of database we use for our distributed databases 
in the oil industry.  We use a set of application software for mapping, 
interpreting seismic reflection data, etc. that is commercially available 
(Landmark is the vendor).  These stand alone applications all use a common 
database interface called Openworks (also a Landmark product) the 
specifications and code for which are made available to us so that we can 
write our own applications as well.  The Openworks program accesses the 
actual databases, both read and write modes, which are Oracle databases 
(another vendor product).  Landmark was using Ingress databases, but for 
reasons unknown to me (probably good ones, I just don't know the reasons) 
they've gone to a large effort to  convert to Oracle databases.  I'm afraid I 
don't know enough about it to tell you the pros and cons of different 
database schemes.

It seems like the analogy with our oil company databases would be Gedcom 
data=database, Lifelines software (or Genserve?)=Openworks interface and 
either Lifelines or unwritten software (e.g. a mapping package) is the 
application.  Does that make any sense? 

Roger Prince


          Friday, October 14, 1994 8:49:42 AM
          GenWeb Item
  From:           Brian
Tompsett,B.C.Tompsett@computer-science.hull.ac.uk,Internet
  Subject:        Announcement of experimental Royal Family Genealogy Web page
  To:             GenWeb
 I have taken the Vic Abell  code for a GENCOM to cgi
conversion tool (which was originated by Frode Kvam )
and modified it to allow access to my large database containing Royal
Genealogical data. My database is larger than other available sets of
similar data, and has been typed in and verified from sources personally.
It currently contains in excess of 7000 names from 5th Century to the present.

 My raw data is stored in Brothers Keeper V5.2 and I modified the original
ged2cgi package to perform the following functions which allow it to display
and search what I have in the base data set:

  - More data fields (Baptism, Accension, Interrment, Divorce etc)
  - Hyperlinks in Notes, comments and sources. This allows the BK5.2
    notation for links to pictures and text files to be maintained.
  - Full support for accented characters. Data in both ISO and PC
    character sets can be supported.
  - extended search capability. The data can be searched based on the
    title or date fields.
  - Indexes can be sorted by Forname. Essential for Royal data when no
    surname is used.
  - Caching of frequent requests to improve access times.

 You are welcome to try browsing the data to see if what I have is a useful
tool. I am currently working on a forms based access method. I am also looking
at the problem of storing endless strings of titles (e.g. Duke of, Duke of, ..)
and endless lists of dates (deposed 1901, reigned 1902-1903, abdicated 1904).

 The modified sources are available at the address listed in the Web page.
The full GEDCOM dataset is not yet available for FTP - it grows by 
hundreds of names a day. My bibliography is available via the web page.
Sources for each data field are not noted or linked to the bibliogaphy;
although respected genealogists might have preferred that!

 My home page is the  
Computer Science Experimental Home Page . The genealogical section can be
accessed directly as  
Experimental GEDCOM to WWW browsing .

 based on this feedback from this restricted list announcement I may announce
more widely (eg roots-l, soc.roots) at some later date.

  Brian Tompsett
PS: please dont hammer my poor computer by frequent and lengthy searches! :-)
PPS: This is just a hobby. :-)


          Friday, October 14, 1994 9:12:58 AM
          GenWeb Item
  From:           jmanley,jmanley@svpal.org,Internet
  Subject:        Documented database
  To:             GenWeb
Is it unreasonable to ask if there could be a database thata 
database 
that is not based on documented fact.  And I realize that documents are 
not necessaryily total fact.  Just my $.02.

Judith<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>


          Friday, October 14, 1994 9:31:37 AM
          GenWeb Item
  From:           jmanley,jmanley@svpal.org,Internet
  Subject:        Re: Documented database
  To:             GenWeb
Something happened to my last post and took my words WAY out of context.

My question is:

Is it possible to have a database that would require documentation?  I 
am aware that even documentation is not always correct.  If not the main 
database "by documentation only", but possibly another one that is?  I 
personally am not interested in greenhorn genealogy.

Judith<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>



          Friday, October 14, 1994 11:03:40 AM
          GenWeb Item
  From:           Brian
Tompsett,B.C.Tompsett@computer-science.hull.ac.uk,Internet
  Subject:        Bug fixed in Royal Date script
  To:             GenWeb
 To the person at 'crpl.cedar-rapids.lib.ia.us' who tried to do a date
search of the Royal database using 
 http://ww.dcs.hull.ac.uk/cgi-bin/geddate/n=royal
 http://www.dcs.hull.ac.uk/cgi-bin/geddate/n=royal?1700-1701
 http://www.dcs.hull.ac.uk/cgi-bin/geddate/n=royal?1953

 *try again* !  
 You found 3 bugs which I fixed as you found them. And thanks.

  Brian
PS: I read the error logs. :-)


          Saturday, October 15, 1994 12:18:20 AM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go Public With Dat
  To:             GenWeb
TO: RIK VIGELAND
FROM: BILL MINNICK
SUBJECT:  Modification of other's databases

In article  rikv@wv.MENTORG.COM (Rik Vigeland) writes:
>   .  .  .  .  .  .  .  
>The point, anyway, is that a small amount of security can go a long way
>(ounce of prevention).  I can quickly scan through any link requests and
>see if they are frivilous hacking or if they require real examination.
>A queue concept allows for a level of quality somewhere between "this is
>great" and "this is trash" to be opened up for review and discussion.

RIK,
   This might be a workable compromise.  I would add that 
I would want to be able to assign access rights to other trusted persons to 
make decisions on the items in my data base's link request queue.   We have  
12 people around the U.S. who are qualified editors on the Richard Austin Data 
Base, any of whom should have the right to act on the queue.  Otherwise I can 
envision severe queue overflow or node shutdown when "Joe Owner" takes an 
extended vacation to Africa.  ( GenServe situation ???)
   Let's plan on assignable access rights to trusted individuals wherever they 
may be located.    -- -- Bill Minnick, Cupertino, CA


          Saturday, October 15, 1994 12:18:30 AM
          GenWeb Item
  From:           John Scoblic,KFJLS@acad1.alaska.edu,Internet
  Subject:        Random thoughts all right.
  To:             GenWeb
Okay so I've been lurking also. I wish I knew how those spiders worked but let 
someone else figure that out.

Concerning additions, changes, pointers. How frustrated have we all been in 
dealing with incorrect data in the LDS system whether the IGI or Ancestral 
File. When there is wrong information or information suspected to be incorrect 
there should be some easy way to look at the source material or contact the 
contributor. Some sources are suspect themselves so who becomes the final 
authority on what is right or wrong and how does a change get effected?

Some discussion with FKC staff leads me to believe one can just submit what 
they believe to be correct information and it will be included in the 
Ancestral File or IGI. But who in Salt Lake decides what info is correct? I 
suspect they just include it in a new record and leave it to others to finger 
out.

So do I want someone linking info to my database? Maybe, maybe not. And who is 
to keep hackers from corrupting the system with crap just cause they figured 
out how to do it. Soon Genweb is more unreliable than the IGI info and becomes 
unuseable. All this work for what?

This is an interesting project but there are tons of questions. Hopefully we 
can find tons of answers.

If this makes no sense let me know and I'll go back to lurking till i 
understand better.

John Scoblic
In rainy, rainy, rainy and windy Ketchikan Alaska.


          Saturday, October 15, 1994 12:18:36 AM
          GenWeb Item
  From:           Roger Prince,trpri@chevron.com,Internet
  Subject:        Changing Data
  To:             GenWeb
Lurker here!

1)  I have some experience as a user and contributor to large databases.  I 
am a geologist for a major oil company.  We are currently in the process of 
converting our large main frame databases to workstation projects.  The 
reason is that the company has become decentralized, but the data hadn't kept 
up.  Hardly anyone uses or contributes to the corporate databases anymore, 
because they are so full of corrupted data and the bureaucracy to change that 
data is so cumbersome and ineffective.  The consequence is that people 
download the data once and then create their own database on a PC (now 
workstations) that they can delete, edit or add to as they work.  The plus is 
the user has absolute control of the data they are working with and that's 
the way we all want it.  The problem of course is that these changes get 
lost, because they don't get put back into any sort of "corporate" archived 
database.  The new scheme is to create project databases which are backed up 
and archived and which are accessible over a unix workstation network.  The 
software we use requires an interpreter tag that is attached to all data 
(seems that GenWeb could use the user's id automatically).  If a new 
interpreter messes with the database, his/her changes are added (filters 
allow you to see only your own, someone else's or all entries).  A user can 
only change those entries for which he/she is the interpreter.  If the user 
wants to change his/her interpretation the database writes over their earlier 
value.  If you really want to preserve both your old and your new entry just 
give yourself a new id.  For genealogical data it seems this would work well 
for all date and name related entries (maybe names would have to use soundex 
codes I don't know?).  Free form notes would still have to be just that. 

2) On a different subject I think it would be nice to tap into the 
geographic/gazeeter servers and link actual lat/long with place names.  Of 
course I'm a map freak, but I'd love to be able to map my ancestors and see 
their distribution in space as well as time.

Thanks for inviting non-programmers to participate!

Roger Prince


          Saturday, October 15, 1994 12:18:43 AM
          GenWeb Item
  From:           Rik Vigeland,rikv@wv.MENTORG.COM,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go Public With Dat
  To:             GenWeb
Chris,

you wrote:

>Also, can we assume that everybody poking at the databases wil be doing so for
>proper reasons?  I can imagine some jerk adding random links to a database for
>no more reason than the fact that he can.  I hope this doesn't happen, but
>people have done worse things to databases and computer systems on the
>Internet already.

>Is the solution to this problem the disallowal person A be from adding links
>between person B's database and person C's database?  Maybe you have to be the
>owner of the database that any link you create points to.

If the genweb has/is a server which accepts commands from anywhere and
everywhere, then consider commands which have access privileges based on
database ownership vs. user (gee, I hope this doesn't mean reinventing UNIX :)

If I am a database 'owner' then I should be able to issue any command which
affects my database.  Other users, who may wish to make a link from my
database to another, or within, would submit changes for approval.  These
changes would go into a queue, which I could peruse at my leisure.  I could
leave an item in the queue indefinitely, approve it (updating the database)
or flush it.  Ideally, the new link would be visible to other users even
while pending approval, and if flushed would return to the submitter.  No
discussion here, yet, on documentation issues or degree of proof quality.

My thinking comes from a vast layer of procedures and approval levels which
we built around a "check-in / check-out" subsystem dealing with schematic
designs in a multi-user environment.  We had seven levels of data quality,
namely three main levels with two levels of review (bidirectional) between
1-to-2, and 2-to-3.  I may still have an X-window diagram of this seven
state machine, if anyone REALLY wants it, but that's probably a bit much
for genealogical data.

The point, anyway, is that a small amount of security can go a long way
(ounce of prevention).  I can quickly scan through any link requests and
see if they are frivilous hacking or if they require real examination.
A queue concept allows for a level of quality somewhere between "this is
great" and "this is trash" to be opened up for review and discussion.

Rik Vigeland
7253 S. Seven Oaks Lane
Canby, OR 97013

Process Automation Engineer
Mentor Graphics Corp.  +1-503-685-1030
Wilsonville, OR

rikv@wv.mentorg.com
also: Prodigy: KTFF04A


          Saturday, October 15, 1994 12:18:49 AM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: Changing Data
  To:             GenWeb
TO: ROGER PRINCE
FROM: BILLL MINNICK
SUBJECT: THANKS FOR SPEAKING OUT;   LET'S HEAR FROM MORE LURKERS

In article  Roger Prince  writes:
>Lurker here!
>1)  I have some experience as a user and contributor to large databases.  I 
>am a geologist for a major oil company.  We are currently in the process of 
>converting our large main frame databases to workstation projects.  The 
>reason is that the company has become decentralized, but the data hadn't kept 
>up.  Hardly anyone uses or contributes to the corporate databases anymore, 
>because they are so full of corrupted data and the bureaucracy to change that 
>data is so cumbersome and ineffective.  The consequence is that people 
>download the data once and then create their own database on a PC (now 
>workstations) that they can delete, edit or add to as they work.  The plus is 
>the user has absolute control of the data they are working with and that's 
>the way we all want it.  The problem of course is that these changes get 
>lost, because they don't get put back into any sort of "corporate" archived 
>database.  The new scheme is to create project databases which are backed up 
>and archived and which are accessible over a unix workstation network.  The 
>software we use requires an interpreter tag that is attached to all data 
>(seems that GenWeb could use the user's id automatically).  If a new 
>interpreter messes with the database, his/her changes are added (filters 
>allow you to see only your own, someone else's or all entries).  A user can 
>only change those entries for which he/she is the interpreter.  If the user 
>wants to change his/her interpretation the database writes over their earlier 
>value.  If you really want to preserve both your old and your new entry just 
>give yourself a new id.  For genealogical data it seems this would work well 
>for all date and name related entries (maybe names would have to use soundex 
>codes I don't know?).  Free form notes would still have to be just that. 
************
ROGER:  Your above commentary presents priceless experience that we must 
factor in to our GenWEB thinking.  
QUESTION:  What database software do you use on your 
work stations, and what specific data base software features must we have to 
implement capabilities which you describe above?
************

>2) On a different subject I think it would be nice to tap into the 
>geographic/gazeeter servers and link actual lat/long with place names.  Of 
>course I'm a map freak, but I'd love to be able to map my ancestors and see 
>their distribution in space as well as time.
***********
ROGER:  Another excellent suggestion.  I have always wanted to create a visual 
presentation of the progression of one's ancestors on a map to the point of 
one's birth.  This would certainly facilitate such a presentation.
***********
>Thanks for inviting non-programmers to participate!
***********
Don't thank us; let us thank you.    -- -- Bill Minnick, Cupertino, CA


          Saturday, October 15, 1994 1:50:40 AM
          GenWeb Item
  From:           baud@research.att.com,Internet
  Subject:        Re: Changing Data
  To:             GenWeb
Roger writes from experience:

> 1)  I have some experience as a user and contributor to large databases.  I 
> am a geologist for a major oil company.  We are currently in the process of 
> converting our large main frame databases to workstation projects.  The 
> ...

If you read this in detail, especially the part about interpreter tags,
owning data, and writing over, you'll see that this is exactly how genserv
works. Score one more for making genweb just a user interface for genserv.

kurt :-)


          Saturday, October 15, 1994 1:54:01 AM
          GenWeb Item
  From:           jmanley,jmanley@svpal.org,Internet
  Subject:        WWW Wishlist
  To:             GenWeb
What I would like to see available is inter-language communication 
capabilities that would enable anyone of any language to commumicate with 
anyone of any language.  EX:  I write in English to someone in Poland and it 
can convert my English to Polish and vice versa.  Perhaps this is not a 
WWW aplication but a software application but Maybe a base relay web for all 
languages.  

With the speaking minds I see on genweb, is this not possible?  Genealogy 
would speed up 100%.  We could have the world on data base in no time.
Judith<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>


          Saturday, October 15, 1994 2:06:31 AM
          GenWeb Item
  From:           T.T.Wetmore,ttw@beltway.att.com,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb
Kurt (>):

>I'm missing something -- why does an index into the databases need to
>be maintained? When it comes time to search, each lifelines database
>could be separately queried, right? As I see it, all you need to do a
>search is a list of databases to search in and query for each one. No
>index need be maintained, just a list of databases.

Kurt,

	I haven't been following the indexing discussions closely enough to
make bright statements, but that's never stopped me from making statements.
Fact.  LifeLines only indexes by name; not by dates, places, and so forth.
When the GENWEB experts mention indexes I think of them as ASCII files (and
therefore "just" LifeLines program outputs) that hold, possibly,
the names of all persons in the database, the dates and places of all
significant events in those persons' lives, keyed, say, to their internal
key values, and so forth.  This index file would be searched first, say to
look for persons matching names, dates and places (LL can only search by
name).  The results of the searching would be, in my opinion, a list of
database keys for persons who should either have their HTML files extracted
and generated, or have specific reports and charts generated for.  These
steps would be accomplished by having LifeLines take the key values and
generate the outputs using the normal report generation feature.  Thus the
index lookup step is "outside" LifeLines, but uses index files generated by
LifeLines.  Hope that makes sense and is in keeping with what the GENWEB
experts have in mind.  If interest I could prototype an index generating
program to what folk think.

			Tom Wetmore


          Saturday, October 15, 1994 2:15:37 AM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: Changing Data
  To:             GenWeb
In article  Kurt Baudendistel  writes:
>Roger writes from experience:

>> 1)  I have some experience as a user and contributor to large databases.  I 
>> am a geologist for a major oil company.  We are currently in the process of 
>> converting our large main frame databases to workstation projects. 

>If you read this in detail, especially the part about interpreter tags,
>owning data, and writing over, you'll see that this is exactly how genserv
>works. Score one more for making genweb just a user interface for genserv.

KURT:  Then would someone among us get a clone of GENSERV running and let us 
evaluate it first hand.  Is the software available for several of us to set up 
pilot data bases for evaluation of the concept?   If not what will it take to 
get it? Is Cliff Manis the developer/ owner,   and is he willing to pass the 
software along? -- -- Bill Minnick, Cupertino, CA


          Saturday, October 15, 1994 5:49:45 AM
          GenWeb Item
  From:           T.T.Wetmore,ttw@beltway.att.com,Internet
  Subject:        LifeLines HTML Program
  To:             GenWeb
Quick report.  I had never looked at an HTML file before, but I took a
GEDCOM record and the corresponding HTML file from the Austin tar file, and
it took about 20 minutes to write a LifeLines program to duplicate the HTML
file.  I had assumed that the database to HTML step would be trivial; it is.

Tom Wetmore, ttw@beltway.att.com


          Saturday, October 15, 1994 12:12:34 PM
          GenWeb Item
  From:           Gene Stark,starkhome!gene@sbstark.cs.sunysb.edu,Internet
  Subject:        My GEDCOM to HTML translator
  To:             GenWeb
I made MS-DOS and MS-windows executables of my GEDCOM to HTML translator
(the one Birger is using in his demonstration).  The entire package,
including C/YACC source code, a copy of the draft GEDCOM 5.3 standard,
and a copy of the "royal92.ged" data, is available at URL:

  ftp://cs.sunysb.edu/pub/TechReports/stark/HTML/ged2html.tar.gz

The program runs rather more slowly under DOS and WINDOWS than it does
under Unix.  In particular, the parsing takes about twice as long,
and the creation of the output files takes about five times as long.
To process large GEDCOMs, you must use the WINDOWS version of the program.
The DOS version of the program can be used for small GEDCOMs only, due
to the limited memory easily available under DOS.

Comments, suggestions, or (especially) copies of troublesome GEDCOMs
output by standard genealogical software packages, are gratefully
accepted.

							- Gene Stark


          Monday, October 17, 1994 7:33:27 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Re: WWW Wishlist
  To:             GenWeb
In message , jmanley typed
:
> What I would like to see available is inter-language communication 
> capabilities that would enable anyone of any language to commumicate with 
> anyone of any language.  EX:  I write in English to someone in Poland and it 
> can convert my English to Polish and vice versa.  Perhaps this is not a 
> WWW aplication but a software application but Maybe a base relay web for all 
> languages.  
> 
> With the speaking minds I see on genweb, is this not possible?  Genealogy 
> would speed up 100%.  We could have the world on data base in no time.
> Judith<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
+ >>>

 ~Sigh~ What you're talking about is a very very difficult problem.

Just last week I attended a "brainstorming session" on a study which is being 
called "multilingual/multicultural computing environments".  It's still rather 
fuzzy just what the study will include, but suffice to say that they expect to 
spend plenty of money just to study the problem; not to solve it.  If the 
study goes well, then they'd like to spend more money to solve the problem.

I'm sure it will happen, but if genweb does it, it will be done by borrowing 
work being done by other people.

Chris


          Monday, October 17, 1994 7:37:50 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Re: lifelines (tracking public changes to data bases)
  To:             GenWeb
In message <9410140353.AA00383@beltway.att.com>, T.T.Wetmore typed:
> Kurt (>):
> 
> >I'm missing something -- why does an index into the databases need to
> >be maintained? When it comes time to search, each lifelines database
> >could be separately queried, right? As I see it, all you need to do a
> >search is a list of databases to search in and query for each one. No
> >index need be maintained, just a list of databases.
> 
> Kurt,
> 
> 	I haven't been following the indexing discussions closely enough to
> make bright statements, but that's never stopped me from making statements.
> Fact.  LifeLines only indexes by name; not by dates, places, and so forth.
> When the GENWEB experts mention indexes I think of them as ASCII files (and
> therefore "just" LifeLines program outputs) that hold, possibly,
> the names of all persons in the database, the dates and places of all
> significant events in those persons' lives, keyed, say, to their internal
> key values, and so forth.  This index file would be searched first, say to
> look for persons matching names, dates and places (LL can only search by
> name).  The results of the searching would be, in my opinion, a list of
> database keys for persons who should either have their HTML files extracted
> and generated, or have specific reports and charts generated for.  These
> steps would be accomplished by having LifeLines take the key values and
> generate the outputs using the normal report generation feature.  Thus the
> index lookup step is "outside" LifeLines, but uses index files generated by
> LifeLines.  Hope that makes sense and is in keeping with what the GENWEB
> experts have in mind.  If interest I could prototype an index generating
> program to what folk think.

Tom,

I'm not certain that the indexes will actually be ASCII files, but 
conceptually, I think they can be viewed that way.  The idea I had is that 
there'd be some way to query a server for it's "index" which wouold be pure 
ASCII, and then the "index server" would take that ASCII data and "do the 
appropriate thing" with it.  Those who query the server wouldn't have to know 
what the "appropriate thing" was.

Chris


          Monday, October 17, 1994 7:51:04 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Re: PAF vs. GEDCOM (was Ancestral File Problems)
  To:             GenWeb
In message , Brian Randell typed:
> I would just add that Genweb should, if possible, be made to act as a front
> end not just to one data repository or type of repository but rather to as
> many worthwhile repositories of genealogical data as possible. The "owners"
> of these repositories and the data that is held in them are by no means
> likely to want to merge their data, or convert to using a common system,
> but a good interface can shield users from the differences between the
> systems, and ideally even the fact that multiple systems exist. (Indeed I'd
> suggest that one of the very important resons for the success of WWW was
> the fact that the people at CERN who designed it thought very deeply about
> how to maximise the utilisation of existing systems, and minimize the
> amount of code that has to be implemented to create WWW.)

(Gee, I disappear for one work day, and the list goes nuts . . . good.)

I agree with this completely.  If we look at how most network resources are 
defined, the protocol definitions don't define what runs behind the protocol; 
they merely define what the protocols themselves look like.  I don't think any 
design should *assume* that lifelines is behind the server, but I also think 
that lifelines shoujld make a good engine for an initial implementation.

Here's a suggestion for the direction we should go:

1) select useful tools for genweb prototyping.  This presumably includes 
lifelines as the general database engine.  It may include listserve for 
managing the email lists that we've been talking about having behind the web 
pages.  It also includes the various gedcom to html scripts that people have 
written as well as a lot of unwritten software.  This set of software should 
grow and change and as people develop or modify software, this list is the 
best way to make sure that everybody knows what tools are available.

2) Using the tools from step one, prototype and try out various ideas.  
Probably we can solve most of what we're trying to accomplish with these tools 
and the WWW protocols.

3) Write a formal specification based on the prototypical work done in step 2. 
 We will then have a reference implementation based on whatever worked best, 
but by formalizing the spec instead of just using what we threw together, we 
have a better chance of being able to run the system on other (non-Unix) 
systems.

Chris



          Monday, October 17, 1994 8:02:57 AM
          GenWeb Item
  From:           Chris Garrigues,cwg@mcc.com,Internet
  Subject:        Re: Modification of other's databases (was Re: lifelines (Why
Go
  To:             GenWeb
In message , Amelia 
Painter typed:
> In reference to Chris's above noted message (one of the few I've 
> understood enough to make a comment on)....as a family history researcher 

Thanks.  Personally, I want the traffic on this list to be both technical as 
we work on implmentations and also philosophical, so what we implement turns 
out to be useful.  There's no reason that anybody with interest in genweb 
can't participate in that discussion.

> I want to know WHERE data comes from and HOW I can verify it.  I was sent 
> down many an incorrect path in the early days of my search for long-death 
> relatives!  Heresay is a nice starting point, but I want to know where I 
> can write or go to obtain an actual copy of any document used in making a 
> connection to my family line.

Do you have any thoughts on how you would like to access this data in a web 
page?

Should there simply be a hypertext link to "References"?  What would you like 
to see on the references page?  Do we need a quick way to telling how good the 
sources are?  Maybe something you can eyeball quickly to tell if the 
references are "hearsay", in paper form in the owners filecabinet, or 
"digitized and actually viewable online.

Unfortunately, most of the sources that I have exists "in my Uncle George's 
file cabinet".  He's promised to will it to me, but (hopefully) that won't be 
for a long time.  I've done the best I can to figure out what his sources are 
from the notes that he's given me, and I'm flushing out my data as I can.  
Unfortunately, this means in some cases my sources are simply a pointer to my 
uncle's email address.

Chris


          Monday, October 17, 1994 9:12:59 AM
          GenWeb Item
  From:           Brian Randell,Brian.Randell@newcastle.ac.uk,Internet
  Subject:        Re: PAF vs. GEDCOM (was Ancestral File Problems)
  To:             GenWeb
Chris:

>I agree with this completely.  If we look at how most network resources are
>defined, the protocol definitions don't define what runs behind the protocol;
>they merely define what the protocols themselves look like.  I don't think any
>design should *assume* that lifelines is behind the server, but I also think
>that lifelines shoujld make a good engine for an initial implementation.
>
>Here's a suggestion for the direction we should go:
>
>1) select useful tools for genweb prototyping.  This presumably includes
>lifelines as the general database engine.  It may include listserve for
>managing the email lists that we've been talking about having behind the web
>pages.  It also includes the various gedcom to html scripts that people have
>written as well as a lot of unwritten software.  This set of software should
>grow and change and as people develop or modify software, this list is the
>best way to make sure that everybody knows what tools are available.

Might I suggest a preliminary step, namely that of briefly investigating
the technical (as opposed to political) problems of front-ending a number
of existing services (Ancestral File, GenServ, RSL, ???) by a common Genweb
interface. Some of these might involve interfacing with Gedcom, others
might not.

Cheers

Brian


          Monday, October 17, 1994 3:07:19 PM
          GenWeb Item
  From:           Gary Hoffman,ghoffman@ucsd.edu,Internet
  Subject:        Re: Changing Data
  To:             GenWeb
All Posters,
Please, in the interest of netiquette and preservatrion of network
bandwidth, please keep quoted material to a minimum.
Thanks,
Gary


          Monday, October 17, 1994 10:38:03 PM
          GenWeb Item
  From:           T.T.Wetmore,ttw@beltway.att.com,Internet
  Subject:        Update on Gengine Program
  To:             GenWeb
Just a quick update.  I've been working on a LifeLines family program that
I'm now calling Gengine.  It has the following command line:

	gengine database report output

where:
	database is the name of a LifeLines database
	report is the name of a LifeLines report program
	output is the name of an output file

The program opens the database and the report program and then runs the
report program.  Any input required by the program is taken from standard
input.  I intend the program to be run without user interaction, which puts
some restrictions on the builtins from the report generation language that
can be used in the Gengine report programs.  For example, in LifeLines itself,
if you use the getindi() (get a person) builtin, and you specify a name
that resolves to more than one person, LifeLines will ask you to select the
correct person.  In Gengine I am assuming that such interaction should not
be possible (please don't tell me how to get around this by simply putting
the correct interaction codes in the program input file -- I know this but
I do not like it), so I now require the response to resolve to exactly one
person.  The only way to do this in general is to specify a key instead of
a name.  But if Gengine is being used in a GENWEB application, all accesses
can be done by key, and in fact, should probably be done that way anyway.
In short this means that I must both change the semantics of a few of the
builtins, and I must leave a few out (eg, all the choose* builtins).
Actually is not a change in semantics so much as a removal of flexibility.

One result of this would be that all report programs runnable by Gengine
are also runnable by LifeLines, but not the converse.

The program is running now, but I am fine tuning the builtins.

Those of you following the source code will be interested in the
redistribution of routines to different files and libraries.  This exercise
has contributed markedly to the modularity of LifeLines code.  The next
LifeLines "family" program will be even easier to write.  Anybody got any
suggestions?

Tom Wetmore, ttw@beltway.att.com


          Tuesday, October 18, 1994 1:21:49 AM
          GenWeb Item
  From:           Herbert
Stoyan,hstoyan@faui80.informatik.uni-erlangen.de,Internet
  Subject:        State of the experiments
  To:             GenWeb
I did not have time to read the mail carefully. So I'm not up-to-date what the
experiments have delivered. I saw at Brian Tompsett's home page a list of
various experiments. All (Brians, Abell, Bratt, Helm, Ladewig) seem to be one
sort of experiment -- work with ged2cgi.c. This does not use lifelines.
Are there experiments with lifelines?
What have they resulted in?

For me, the genweb is not a web of unconnected sites. Between Brians royal
Files and my German nobility index there's much interrelation. So there must
be links between both sites. Is ged2cgi.c or the lifelines-experiments able
to handle that?

Is somebody who did compare the solutions? What are the pros and cons?

I would like to conevrt my system into one of the solutions and we should
propose the same to newcomers (even if the solutions have to be cleaned up).
Herbert


          Wednesday, October 19, 1994 11:57:03 PM
          GenWeb Item
  From:           Gene Stark,starkhome!gene@sbstark.cs.sunysb.edu,Internet
  Subject:        Re: My GEDCOM to HTML translator
  To:             GenWeb
>GENE:
>   My next project on GenWEB is to put out our 6800-person Richard Austin 
>Descendants data base for viewing, comments and general use.  I want to put 
>this data base onto my provider's Sun work station running under SunOS.  I 

>   Can I easily configure your GEDCOM-HTML software to operate this way on the 
>Sun Machine with a current copy of LifeLines and a GEDCOM file of our Richard 
>Austin data base ?   At this time, I am not on-line full time with my PC, and 

Sounds like Birger is the one to answer that one.  I haven't yet gotten
around to try LifeLines and his setup, so I don't know about the ease of
configuring it.  I got the impression from my communications with him, though,
that he was running on a Sun Sparc, so depending on what kind of Sun you
have, your host environment might be identical to his.  This would likely
make it easier to set up.
							- Gene Stark



          Wednesday, October 19, 1994 11:57:06 PM
          GenWeb Item
  From:           Brian
Tompsett,B.C.Tompsett@computer-science.hull.ac.uk,Internet
  Subject:        Results from my experiment with a Royal database
  To:             GenWeb
At present the GENWEB list is 'write-only' for me. I do not see anything posted
to it! This is because the ucsd machine cannot seem to see my machine for
delivery of mail, but my machine is happy to see ucsd! I expect it will get
solved in time, as these things usually do.

 I say this because, you might be having some neat discussion and have already
solved the problems I encountered; I have no way of knowing or seeing.

 I have had feedback from a few people from my earlier contribution, and I 
thank them for it. I have put the essence of these comments into a disussion
document entitled 
Limitations of the GEDCOM to Web Experiment. (I am of course making a
niave assumption that people have Mime/html aware mail readers that can
follow this link or can at least cut and paste into a www/mosaic open. If
not you can ftp from the same address/path. )

  Brian
(who doesnt have any spare time to do this but does anyway)



          Wednesday, October 19, 1994 11:57:08 PM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: My GEDCOM to HTML translator
  To:             GenWeb
In article  starkhome!gene@sbstark.cs.sunysb.edu (Gene Stark) writes:
>I made MS-DOS and MS-windows executables of my GEDCOM to HTML translator
>(the one Birger is using in his demonstration).  The entire package,
>including C/YACC source code, a copy of the draft GEDCOM 5.3 standard,
>and a copy of the "royal92.ged" data, is available at URL:

>  ftp://cs.sunysb.edu/pub/TechReports/stark/HTML/ged2html.tar.gz

GENE:
   My next project on GenWEB is to put out our 6800-person Richard Austin 
Descendants data base for viewing, comments and general use.  I want to put 
this data base onto my provider's Sun work station running under SunOS.  I 
want to use Lifelines and the GEDCOM-HTML converter exactly as Birger Wathne 
is using it on his Royalty line.  From my home page, I want to have a name 
Index page with URL's which activate Lifelines to deliver up an HTML page on 
the requested individual.
   Can I easily configure your GEDCOM-HTML software to operate this way on the 
Sun Machine with a current copy of LifeLines and a GEDCOM file of our Richard 
Austin data base ?   At this time, I am not on-line full time with my PC, and 
therefore I believe I can not use a PC version of your software.
    I also welcome advice from Tom Wetmore and/or Birger Wathne which will let 
me get the Richard Austin data base,  a large, well source-documented data 
base on line the fastest, easiest way possible.
    Regards,   Bill Minnick, Austin Computer Project, Austin Families of 
America Association


          Thursday, October 20, 1994 12:18:29 AM
          GenWeb Item
  From:           T.T.Wetmore,ttw@beltway.att.com,Internet
  Subject:        Re: Update on Gengine Program
  To:             GenWeb
John Chandler, 

>One of the early principals of UNIX program argument design was not to
>name output files as ordinary arguments.

Good point.

>...why not just have the report output go to STDOUT, with any commentary
>on STDERR?  Then there would be no need for either an argument or an
>option ...

Another good point.

>>... so I now require the response to resolve to exactly one person.  The
>>only way to do this in general is to specify a key instead of a name.

>This is not a good solution.  If you want to generate a book or chapter
>starting at a given person, you need to have a way to unambiguously
>specify that person.  If you use a key, you're out of luck when you
>reload the database at the next LifeLines release.

The intention of Gengine is to run as a standalone server in a GENWEB
application.  In such a context, I am assuming that a URL process will be
controlling the interaction based on user actions taken while working with
HTML files.  The links within the HTML files will, almost certainly, be
encoded with keys, not names, so identification via key seems the natural
approach.

>Previously I've suggested having a LifeLines primitive that permits
>specifying a person by name and a known date, such as birth or death
>date.  Wouldn't cover all cases, but most.  Would this not be possible?

>>But if Gengine is being used in a GENWEB application, all accesses
>>can be done by key, and in fact, should probably be done that way anyway.

>Unless a user wants to jump to a person by name, rather than following links.

I don't think so.  Though the person viewing the HTML files will see and
click on names, the link coding scheme, that the user does not see, will be
based on keys.  It would be good to get confirmation on this from a WWW and
HTML type person.  Note that LL will soon support the concept of a user
reference key, a user assigned, permanent key, that will solve the problem
you mention.

Tom Wetmore, ttw@beltway.att.com


          Thursday, October 20, 1994 8:01:08 PM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,World
  Subject:        Re: Documented database
  To:             GenWeb
In article  jmanley  writes:
>Is it possible to have a database that would require documentation?  I 
>am aware that even documentation is not always correct.  If not the main 
>database "by documentation only", but possibly another one that is?  I 
>personally am not interested in greenhorn genealogy.

Judith:
    We had the same concern 5 years ago at our Silicon Valley PAF Users Group, 
so we set out to write a guideline for documenting sources for individual 
records which are included in data bases.  We have published this quide each 
year for the past 4 years; the Jan 1994 version is availableby anonymous ftp 
as follows:

ftp.rahul.net/pub/svpafug/dguidtxt.txz             ascii, GZIP version 
ftp.rahul.net/pub/svpafug/dguidtxt.zip             ascii, pkZIP version
ftp.rahul.net/pub/svpafug/docguid4.zip           WINWORD 2.0c, pkZIP version

We'll be putting out a 6800 person data base on the WWW which has been 
documented to this guideline.  We offer this guideline as a starting point for 
establishing GenWEB documentation guidelines.

Regards,

Bill Minnick, VP Silicon Valley PAF Users Group



          Thursday, October 20, 1994 9:16:21 PM
          GenWeb Item
  From:           T.T.Wetmore,ttw@beltway.att.com,World
  Subject:        Re: Update on Gengine Program
  To:             GenWeb
Bill,

>I would dearly like to get that Richard Austin Data BAse on line using 
>Lifelines and some scheme for dynamic creation of HTML pages rather than 
>working with 6800 individual HTML files.  Can I read between tbe lines
>in your last several GENWEB items that you are close to being able to do
>this ???

I have the piece parts that allow a WWW capable person to do this.

>If you can send me an operating set of Lifelines software and GEDCOM-HTML
>routine, and a way to interpret URL's, I should be able to take it from
>there and put the software up on the Sun eqpt (SunOS) at a2i, my internet
>provider.  I would hope you can get this all going at your end to check it
>out before sending it along.  

I usually distribute LifeLines by email with simple instructions on how to
build it included.  I'll send it to you when I'm done with this note.
Building for SUNs is never a problem.  I will have to write the GEDCOM to
HTML routine again, as the first one I wrote is now on my UNIX PC that
won't boot.  I don't know yet how to interpret URL's, so I can't help you
yet on that part.

Note.  I have a friend who has recently set up a Mosaic system using
LifeLines, a GEDCOM to HTML LifeLines extractor, and a URL setup, that
does, I imagine, exactly what you and other WWW explorers are thinking
about.  I may get to see him next Tuesday, and if I do, I'll ask him to
explain his set up to me.

Tom Wetmore


          Sunday, October 23, 1994 3:36:12 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Re: My GEDCOM to HTML translator
  To:             GenWeb
----------
X-Sun-Data-Type: text
X-Sun-Data-Description: text
X-Sun-Data-Name: text
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 54


First, I have a CGI gateway in my http-daemon. This is a simple script
that does some rudimentary syntax checking, and then runs another script
that does the actual database searching and HTML converting. The reason
for this is that I have the httpd running on a Solaris 1.1.1 host, while
lifelines and ged2html run on a Solaris 2.3 host, so my first script
rsh's to the second host. These two scripts can easily be combined into one.
I'll do so myself once I upgrade the server to 2.3.

You then need Gene's ged2html and ttw's lifeline engine. Mail him
directly to get the source for 3.0.1, as you'll want the read-only option.

I don't actually run 3.0.1 yet. I don't have the time to convert the
setup, even if it should be only a few minutes of work. Too busy.....

I'll append the two scripts I use as Sun mailtool attachments. You should
be able to extract them without Sun mailtool as well, as there are nice
ascii headers with filename and data format.

llgwrun will leave all temporary files in /tmp until you remove the comment
in front of the rm line in the script. Remove this comment after initial testing.

I should develop the script so it mails me any error messages from ged2html
along with index number and database information, so I would get automatic
information about failed lookups.


Birger


>From list-relay@UCSD.EDU  Wed Oct 19 04:46:33 1994
>Date: Tue, 18 Oct 1994 18:53:20 -0400
>From: starkhome!gene@sbstark.cs.sunysb.edu (Gene Stark)
>To: svpafug@rahul.net (Bill Minnick)
>Cc: genweb@UCSD.EDU
>Subject: Re: My GEDCOM to HTML translator
>
>>GENE:
>>   My next project on GenWEB is to put out our 6800-person Richard Austin 
>>Descendants data base for viewing, comments and general use.  I want to put 
>>this data base onto my provider's Sun work station running under SunOS.  I 
>
>>   Can I easily configure your GEDCOM-HTML software to operate this way on the 
>>Sun Machine with a current copy of LifeLines and a GEDCOM file of our Richard 
>>Austin data base ?   At this time, I am not on-line full time with my PC, and 
>
>Sounds like Birger is the one to answer that one.  I haven't yet gotten
>around to try LifeLines and his setup, so I don't know about the ease of
>configuring it.  I got the impression from my communications with him, though,
>that he was running on a Sun Sparc, so depending on what kind of Sun you
>have, your host environment might be identical to his.  This would likely
>make it easier to set up.
>							- Gene Stark
>
----------
X-Sun-Data-Type: shell-script
X-Sun-Data-Description: shell-script
X-Sun-Data-Name: ll-gw
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 77

#!/bin/sh
#
# ll-gw : WWW to lifelines gateway
#
# CGI gateway program for NCSA httpd and LifeLines
#
# by Birger A. Wathne, Skrivervik Data AS, 1994
# Birger.Wathne@sdata.no
#
#
# Supported lookup types:
#
# Index - Generate index of database
# Search - Search form
# LookupInternal - Look up person based on internal reference ID
# Lookup - Look up person based on stable reference ID
#
# Only Lookupinternal has been implemented. The other options
# will be implemented when support is available in LifeLines (real soon now)
#

if [ $# -ne 1 ]; then
	echo "Content-type: text/plain"
	echo
	echo "Argument count error"
	exit
fi

LLSERVER=morgan
LLRUN=/net/data/lines/bin/llgwrun
LLDATABASE=`echo $PATH_INFO|sed -e 's/^.*BASE=\([^\/]*\).*$/\1/'` 
LLINDEX=`echo $PATH_INFO|sed -e 's/^.*INDEX=\([^\/]*\).*$/\1/'` 

case $1 in
"LookupInternal" )
	rsh $LLSERVER $LLRUN $1 $LLDATABASE $LLINDEX 

	exit
	;;

"Search" | "Index" | "Lookup")
	rsh $LLSERVER $LLRUN $1 $LLDATABASE
	exit
	;;

"Test" )
	echo "Content-type: text/plain"
	echo
	echo argc is $#. argv is "$*".
	echo
	echo LLDATABASE = $LLDATABASE
	echo LLINDEX = $LLINDEX

	echo SERVER_SOFTWARE = $SERVER_SOFTWARE
	echo SERVER_NAME = $SERVER_NAME
	echo GATEWAY_INTERFACE = $GATEWAY_INTERFACE
	echo SERVER_PROTOCOL = $SERVER_PROTOCOL
	echo SERVER_PORT = $SERVER_PORT
	echo REQUEST_METHOD = $REQUEST_METHOD
	echo HTTP_ACCEPT = "$HTTP_ACCEPT"
	echo PATH_INFO = "$PATH_INFO"
	echo PATH_TRANSLATED = "$PATH_TRANSLATED"
	echo SCRIPT_NAME = "$SCRIPT_NAME"
	echo QUERY_STRING = "$QUERY_STRING"
	echo REMOTE_HOST = $REMOTE_HOST
	echo REMOTE_ADDR = $REMOTE_ADDR
	echo REMOTE_USER = $REMOTE_USER
	echo AUTH_TYPE = $AUTH_TYPE
	echo CONTENT_TYPE = $CONTENT_TYPE
	echo CONTENT_LENGTH = $CONTENT_LENGTH
	exit
	;;
esac

echo "Content-type: text/html"
echo
echo "Illegal argument"
----------
X-Sun-Data-Type: default
X-Sun-Data-Description: default
X-Sun-Data-Name: gedreport
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 25

proc main ()
{
	indiset(s)
	indiset(t)
	indiset(a)
	getindi(i, "Enter person to center GEDCOM around")
	addtoset(s, i, 0)
	set(a, childset(s))
	set(t, spouseset(s))
	set(a, union(a, t))
	set(t, parentset(s))
	set(a, union(a, t))
	set(a, union(a, s))
	"0 HEAD" nl()
	"1 SOUR LIFELINES" nl()
	"2 VERS 3.00b" nl()
	"1 SUBM @S0@" nl()
	"1 GEDC" nl()
	"2 VERS 5.2" nl()
	"0 @S0@ SUBM" nl()
	"1 NAME Birger /Wathne/" nl()
	gengedcom(a)
	"0 TRLR" nl()
}

----------
X-Sun-Data-Type: shell-script
X-Sun-Data-Description: shell-script
X-Sun-Data-Name: llgwrun
X-Sun-Charset: us-ascii
X-Sun-Content-Lines: 61

#!/bin/sh
#
# llgwrun : WWW to lifelines gateway backend
#
# CGI gateway program for NCSA httpd and LifeLines
#
# by Birger A. Wathne, Skrivervik Data AS, 1994
# Birger.Wathne@sdata.no
#
#
# Supported lookup types:
#
# Index - Generate index of database
# Search - Search form
# LookupInternal - Look up person based on internal reference ID
# Lookup - Look up person based on stable reference ID
#

case $1 in
"LookupInternal" )
	if [ $# -ne 3 ]; then
		echo "Content-type: text/plain"
		echo
		echo "Argument count error"
		exit
	fi

	echo "Content-type: text/html"
	echo

	mkdir /tmp/ged.$$
	cd /tmp/ged.$$
	LLDATABASES=/net/data/lines/db
	LLPROGRAMS=/net/data/lines/reports
	LLREPORTS=/tmp/ged.$$
	TERM=vt100
	export LLDATABASES LLPROGRAMS LLREPORTS TERM

	index=`echo $3|cut -c2-`
	( echo rgedreport ; echo $index ; echo ireport.ged ; echo q ) |
/net/apps/fhf/lifelines/lines3/bin/lines300b $2 > /tmp/ged.$$/logfile 2>&1
	/netopt/ged2html/bin/ged2html -s $3 -u
"http://www.vest.sdata.no/cgi-bin/ll-gw/BASE=$2/INDEX=%s?LookupInternal" -f
"
Demonstration of link between LifeLines and WWW by Birger
Wathne" < /tmp/ged.$$/report.ged > logfile 2>&1
	cat /tmp/ged.$$/$3.html
	cd /
#	rm -rf /tmp/ged.$$

	exit
	;;

"Search" | "Index" | "Lookup")
	echo "Content-type: text/plain"
	echo
	echo "Unimplemented option"
	echo "Search strings = $PATH_INFO"
	exit
	;;

esac

echo "Content-type: text/html"
echo
echo "Illegal argument"



          Sunday, October 23, 1994 8:20:22 AM
          GenWeb Item
  From:           Bill Minnick,svpafug@rahul.net,Internet
  Subject:        Re: Request Your Help Getting Richard Austin Data Base On Line
  To:             GenWeb
In article  Birger.Wathne@vest.sdata.no (Birger A. Wathne) writes:
>The binaries for ged2html and lifelines will have to be recompiled
>unless your provider runs Solaris 2.3. They should be easily portable to any
>UNIX system. My scripts should also work on any UNIX. As you can see, I
>haven't really done much work, just thrown together existing utilities,
>and fired off this discussion. I don't even have the time to follow
>the discussion closely anymore...


Birger:
    I truly appreciate you making your initial GenWEB solution available,  
 especially in light of your other commitments.  I will do my best to build on 
your work and help others do the same.  Thank you for lighting the way.  My 
challenge will be to find the time to put your efforts to work. 

Regards,

Bill MInnick



          Tuesday, October 25, 1994 10:25:45 AM
          GenWeb Item
  From:           Herbert
Stoyan,hstoyan@faui80.informatik.uni-erlangen.de,Internet
  Subject:        Birgers programs
  To:             GenWeb
It was a good idea to post the solution, but not, to send binaries around. I
would like to adapt them tou my configuration.

How to install the http-daemon?


          Tuesday, October 25, 1994 11:47:28 AM
          GenWeb Item
  From:           Everton,jayhall@xmission.com,Internet
  Subject:        New WWW page for genealogists
  To:             GenWeb
 
A new World-Wide Web page for genealogical resources on the Internet is now 
available, sponsored by Everton Publishers. If you have a Web browser, 
point it to:
 
                     http://www.xmission.com/~jayhall/
 
for the front page, with links to over 200 files, Web pages, gopher menus, 
and databases. Current areas include: Getting Started With Your 
Genealogical Research, Genealogical Resources in the United States, 
Non-U.S. Genealogical Resources, Special Genealogical Resources, Using the 
Internet, and Everton Publishers, with other areas currently under 
construction.
 
--------------------------------------------------------------------------
Jay Hall, Everton Publishers     | WWW = http://www.xmission.com/~jayhall/
165 South West Temple, Suite 200 | E-Mail = jayhall@xmission.com
Salt Lake City, UT 84101         |



          Wednesday, October 26, 1994 10:58:35 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Re: Birgers programs
  To:             GenWeb

>It was a good idea to post the solution, but not, to send binaries around. I
would like to adapt them tou my configuration.
>
>How to install the http-daemon?
>

There were no binaries in my posting, only short shell scripts.
So they are completely configurable.

My gateway script works nicely with the NCSA httpd. I have not tested it with
the cern httpd. Just install the ll-gw script in the cgi-bin directory
for the httpd, and edit it to suit your local configuration.

You'll have to get the http daemon itself from NCSA (The same ftp-server
that you get Mosaic from?). I should have included an URL, but I am
supposed to be playing with my band in 10 minutes, so I'll have
to get out of here NOW!



Birger



          Thursday, October 27, 1994 10:23:21 AM
          GenWeb Item
  From:           Todd Fries,tfries@umr.edu,Internet
  Subject:        Programs...
  To:             GenWeb
OK, so I've been a lurker throughout most of the discussion...maybe I haven't
read every single article, but I'm currently running Linux on my machine, 
and I'm just wondering 1 simple question:

That programs are out there to 'enter' the geneological information into?  I'm
looking for not only executables but source, because an uncle of mine wants me
to program a 'graphical' output to a plotter that would display not only who
descended from who, (a descendants chart) but also graphically show the 
life spans of each person, so that a timeline on the side of the paper would
match up with the birth and death of each person...and having the storage   
portion of the program already programmed would GREATLY ease my job...just add
another method of output...

I'm sorry if you have already mentioned what programs, and where to get them,
but I've not caught it when/if you did...                   

So, could I have either URLS to filenames or maybe just files I could archie 
for?  

Thanks,
Todd....tfries@umr.edu


          Friday, October 28, 1994 4:15:56 PM
          GenWeb Item
  From:           Gary Hoffman,ghoffman@ucsd.edu,Internet
  Subject:        Fwd: Genealogy Programs
  To:             GenWeb
Todd Fries (tfries@umr.edu) asked about genealogy data entry programs.

Todd, this list is not the best place to seek information about common
genealogy database programs. There are quite a number for all the popular
hardware platforms. The best ones can interchange information with each
other and other utilities using the GEDCOM format. I myself am interested
in a timeline type display and also a geographical movement display. The
timeline is available in some of the specialized programs and utilities. I
suggest anyone interested in genealogy data entry and display subscribe to
the ROOTS-L list and its progeny. For more information, you should visit
the Genealogy Home Page (http://ftp.cac.psu.edu/~saw/genealogy.html).

Having said that, let me state that any genealogy program can be used for
inputting data into a GenWeb system as long as it outputs a GEDCOM file.
Several folks on this list are working on GEDCOM to HTML conversions and
others are working on the reverse of that, that is, HTML back to GEDCOM.
Maybe we can even use Mosaic/Netscape/Lynx to display your timelines and my
map charts someday.

Cheers,

Gary



          Sunday, October 30, 1994 5:50:26 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Updated GenWeb demo
  To:             GenWeb

I have now upgraded my GenWeb demo a little:

- It now uses Lines 3.0.1 in read-only mode, so the should not
be any problems with concurrent lookups anymore.

- I have added a logo at the top of the pages, and a button bar at the bottom.
Only the 'Info' and 'Help' buttons are operational. The Mail button
opens a form, but you can't actually mail anything with it yet.
I have tested the images on a gray-scale monitor, but not on real
black'n'white. If there are any problems, please tell me so.
I have tested using Mosaic and lynx.


If anyone can suggest really good icons for the Search and Index buttons,
please do so. I have 'borrowed' the mail icon from Sun's OpenWindows,
and made the other icons myself.

The 'GenWeb' logo I use is something I just threw together. I guess we should
design some common logo for the GenWeb project? Any artists out there?

I would like source for:
1) A LifeLines report program to generate indexes. Anything I can use as a
   base for developing a new macro to generate indexes. I'm looking for efficient
   code samples to extract all names or names within a given range, efficient
   sorting of individuals, etc. I have not done much LifeLines programming yet...

2) Search reports for LifeLines. Anything that lets a user interact with
   the report program to extract one or more records that match given
   criteria. As this will run non-interactively, I don't want to use the
   getindi function.



Birger


          Sunday, October 30, 1994 6:09:33 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Updating GenWeb records
  To:             GenWeb

There has been some discussion regarding updating/adding information
to GenWeb records. My answer would be:

If a user discovers that he/she has something to add or change in someone
else's GenWeb database, he/she should approach the data provider, and
offer his/her data. This could be automated by the database
provider in several steps.

Step one would be the one I'm currently implementing in my demo; a
simple form to let user's mail back comments.  I would then have to
decide how to incorporate new info into my database.

Step two could be a more elaborate form, that would allow any user
to add comments directly into the database. Some kind of heading
would be added to indicate who added the info, and when.

If you really trust the world, you could even make a form to
let them add any kind of tags.

The GenWeb should also support the notion of one person having several
URN's around the world. So if you can't come to an agreement with the
other provider, make your own record, and make a link to the
other person's record within your data.

Note that this kind of implementation would work just as well with
mail gateways as with mosaic.



Birger


          Sunday, October 30, 1994 6:44:38 AM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Re: Updated GenWeb demo
  To:             GenWeb

I just added two more fixes:

- Users get an error message when LifeLines fails with error status
(typically when it fails to open the base because I'm updating it...)

- Users get an error message when ged2html fails with error status.
The error log from ged2html should be mailed to me automatically, but I
have not tested this feature yet.


Test it out, and mail me if you want me to post the updated scripts
in a few days.

(My wife has gone to visit my mother for a week, so I have a little
more time for GenWeb this week).



Birger


          Monday, October 31, 1994 3:49:41 AM
          GenWeb Item
  From:           Herbert
Stoyan,hstoyan@faui80.informatik.uni-erlangen.de,Internet
  Subject:        Genweb, ged2html
  To:             GenWeb
I moved my Descendants of Charlemagne to Lfelines and ged2html. Look at:
http://faui80.informatik.uni-erlangen.de/html/chl-enter.html

There are some problems, mainly with ged2html:

I need multi-language versions. Currently, the version puts in annotations like
BIRTH, DEATH, BURIAL, MARRIAGE, DIVORCE, CHRISTENING for which international
signs are in use:
*,+,# (box), oo (infinity), O|O, ~. This should be changed. I tried it but
somehow the changed program does not run.

It should be possible to translate names. There should be a translation table
for firstnames and family names.

I do not like the presentation of family names in capitals. At least, there 
should be a switch to avoid that.

The system enables one to add person relevant info in an Include directory.
This means, there will be a huge directory which mirrors our gedcom files.
This is not good: We need a new format for additional person info.

Additionally, gedcom lacks the following things:
predecessor-successor links (for some office, like imperatorship, kingship etc.)
external-db-links (for people which are mentioned in one db but the full info
is in another db)



          Monday, October 31, 1994 5:27:02 AM
          GenWeb Item
  From:           Herbert
Stoyan,hstoyan@faui80.informatik.uni-erlangen.de,Internet
  Subject:        ged2html
  To:             GenWeb

>>I need multi-language versions. Currently, the version puts in annotations like
>>BIRTH, DEATH, BURIAL, MARRIAGE, DIVORCE, CHRISTENING for which international
>>signs are in use:
>>*,+,# (box), oo (infinity), O|O, ~. This should be changed. I tried it but
>>somehow the changed program does not run.

>Please clarify...  Do you want people to send you translations to be
>included in the software, or are you asking the ged2html maintainers
>to support several languages?

>Regards,
>Anders Andersson, Dept. of Computer Systems, Uppsala University
>Paper Mail: Box 325, S-751 05 UPPSALA, Sweden
>Phone: +46 18 183170   EMail: andersa@DoCS.UU.SE

Gd2html should support several languages.


          Monday, October 31, 1994 2:34:27 PM
          GenWeb Item
  From:           Birger A. Wathne,Birger.Wathne@vest.sdata.no,Internet
  Subject:        Re: Genweb, ged2html
  To:             GenWeb
>
>It should be possible to translate names. There should be a translation table
>for firstnames and family names.
>

I'm not completely shure that this would be ged2html's job. I agree that it
should be possible to change the tags that ged2html adds, but I'm not
shure that it would be ged2html's job to edit the data.

ged2html should use locale functions to get the strings for tags, etc.
This would make it possible for anyone to add new translations without
adding source code. Just set up a file with the new strings for
the program, put the file in the proper directory, set the LC_MESSAGES
variable, and run ged2html.

>I do not like the presentation of family names in capitals. At least, there 
>should be a switch to avoid that.
I agree with this one.

>The system enables one to add person relevant info in an Include directory.
>This means, there will be a huge directory which mirrors our gedcom files.
>This is not good: We need a new format for additional person info.
I never use this option. I always put all my data in the database.

>Additionally, gedcom lacks the following things:
>predecessor-successor links (for some office, like imperatorship, kingship etc.)
GEDCOM is far from perfect :(

>external-db-links (for people which are mentioned in one db but the full info
>is in another db)
I feel that the proper way would be some modification to the semantics of
xref's. But this would involve not only ged2html, but maybe LifeLines as well.
Another alternative is to add a new GEDCOM tag.



Birger


          Monday, October 31, 1994 8:48:22 PM
          GenWeb Item
  From:           Gene Stark,starkhome!gene@sbstark.cs.sunysb.edu,Internet
  Subject:        Re: Genweb, ged2html
  To:             GenWeb
>>It should be possible to translate names. There should be a translation table
>>for firstnames and family names.
>>
>
>I'm not completely shure that this would be ged2html's job. I agree that it
>should be possible to change the tags that ged2html adds, but I'm not
>shure that it would be ged2html's job to edit the data.

I agree with this.

>ged2html should use locale functions to get the strings for tags, etc.
>This would make it possible for anyone to add new translations without
>adding source code. Just set up a file with the new strings for
>the program, put the file in the proper directory, set the LC_MESSAGES
>variable, and run ged2html.

There is a problem with this:  I don't think locale functions are
completely implemented on enough systems to make the code portable.
I already had problems trying to compile the code under MS-DOS due to
to the use of the getopt() library function.  I had to hack a cheap
substitute.  I could add ad hoc support for output in different
languages, though.

>>I do not like the presentation of family names in capitals. At least, there 
>>should be a switch to avoid that.
>I agree with this one.

OK, I'll do something about this one.

							- Gene Stark