USGENWEB-SW-L Archives

From: "David W. Morgan" <>
Subject: [USGENWEB-SW] Re: Fwd: Re: USGW-CC-L: Search Engine Proposal Followup
Date: Tue, 17 Aug 1999 16:40:06 -1000 (HST)

> >Date: Wed, 04 Aug 1999 14:02:22 -0700
> >From: Fred Smoot <>
> >Organization: Dog Trot X Press
> >X-Mailer: Mozilla 3.0Gold (Win95; U; 16bit)
> >To:
> >CC:
> >Subject: Re: USGW-CC-L: Search Engine Proposal Followup
> >Sender:
> >Reply-To:
> >
> >Greetings all,
> >
> >This is in response to John Rigdon's original missive, copied down
> >under. Sorry about the excessive bandwidth usage, folks.
> >
> >Hi John,
> >
> >The main consideration is control. Either the Project controls its own
> >tools or someone else will.
> >
> >Whether we were to place the tools on a "free" commercial server or a
> >tax supported site, the Project's control is far from absolute. Only if
> >the Project faced the challenge to establish its own in-house systems,
> >could it ultimately control its own destiny. While the software part of
> >the equation is easy, the political ramifications are extreme. I doubt
> >that we will ever get to the point of agreement on the in-house
> >political issues. There are just way too many vested interests in our
> >Project.
> >
> >So if we rule out the Project as system owner, we need to look at the
> >other options. While we discuss a search engine for the whole project,
> >we should keep in mind that some states may be able to develop and/or
> >implement their own state tools. This may be a short term expediency for
> >some states. Also, RootsWeb could with its already existing tools,
> >search the all of our Project. They could, but probably won't. It would
> >seem to me that it is not to their commercial advantage to send folks to
> >"off" RootsWeb sites. They need the "hits," not the "misses." Of course,
> >some have hinted that all county sites should be forced to RootsWeb, and
> >if they could accomplished that, all this dialogue would be moot. The
> >CCs would have the tools right at hand at RootsWeb. No off site search
> >would be necessary. Still, we will see what we will see.
> >
> >But back to the main question. If we can't get our internal problems
> >solved and we can't reach a resolution with RootsWeb, then in order to
> >offer the tools and systems we need, we must then look elsewhere. In
> >order to serve our entire Project, we would need to form some kind of
> >understanding with an outside entity or somehow create our own entity.
> >There can be big time problems with outside companies, because companies
> >can change direction on us, they can put unreasonable restrictions on
> >us, or even sell out to a less than friendly entity. Even a tax
> >supported site could be a problem if the public entity went through a
> >budget cut and found it necessary to go on a bandwidth diet.
> >
> >We could also form a separate group called "Friends of Free Online
> >Genealogy" or some such, and we could develop tools and provide them to
> >the Project's "off" RootsWeb states. Since this move would be outside
> >the Project's control, there could be no interference in the development
> >of such a system, and of course, the use of those tools would be
> >optional.
> >
> >This group would be established by ten of the True and Faithful.
> >Membership to this group would be by invitation only. This way, the
> >naysayers and the ringers, the employees of foreign powers with hidden
> >agendas, will be eliminated from participation. No boulders in the road
> >to impede the traffic flow would be tolerated.
> >
> >The ten then get to work on obtaining tools, non-profit incorporation,
> >IRS tax exempt status, taking a mature look at fund raising, plan for
> >future expansion such as USConnect, mailing lists, potentially hosting
> >some independent county sites, and much more.
> >
> >They will also look at the server options. The first being, contracting
> >with some outside server, a neutral third party. Or second, establishing
> >a robust server system of its own. In either case, Friends of Free
> >Online Genealogy becomes the provider of the tools and the data remains
> >in the ownership of the CCs. The AUP would read simply, What is yours,
> >is yours, we only provide the tools. Period. End of paragraph.
> >
> >If the decision was made to develop a robust server system, then we
> >could start of off with perhaps a couple of new Sun servers, and a T3
> >line, although we could go on the cheap and use multiple T1 lines. We
> >would require a physical plant location. Aside from the real money
> >issues, the location becomes the single major issue. We *not* want a
> >location that is in some shack on an obscure mountain top. Additionally,
> >the location must be accessible to the maintainer of the equipment. The
> >maintenance is also a cost issue, but I have in mind a short term work
> >around for that. In time, we would need pay the for the maintenance. We
> >would need a salaried technician or two.
> >
> >We would set up a system that pays NO salaries except to the
> >technicians. We would focus simply on providing free tools to all the
> >independent folks in our Project. We would eliminate the ego and control
> >issues that are so common within the Project. We would have the support
> >of those that do not want to be controlled by others or have their work
> >redistributed without their consent.
> >
> >In closing, remember that this effort would start with a selected
> >altruistic few and it would avoid Project politics and those with
> >outside interests. Food for thought, John?
> >
> >Fred Smoot
> >
> >****
> >
> > wrote:
> > >
> > > Since I am receiving this list in digest mode, it allows me to read and
> > > consider several responses to my thoughts, before I respond.
> > >
> > > The ideas proposed so far are valid and I admit I don't have the answers to
> > > all of the questions. One aspect I did not clarify however is the initial
> > > startup costs and implementation of the search engine. I essentially have
> > > that in place, and I do not believe any significant funding would be
> > required
> > > to implement the search engine. The costs will come with the on-going
> > > maintenance and band width issues which I have not carefully documented
> > yet.
> > >
> > > That is a moot point though in view of the idea that current cc's may not
> > > want to participate if the venture is commercial in any nature. If the
> > issue
> > > really boils down to no commercial activities whatsoever, then we must
> > either
> > > (1) resign ourselves to the fact that no "piece" of the project can be
> > larger
> > > than what any individual volunteer is able to fund, or (2) we must find a
> > > government or non-profit organization to fund the search engine on a grant
> > > basis.
> > >
> > > I guess there's a third option - we can grouse about the current situation,
> > > but essentially learn to live with it and allow the current entities to
> > > continue to develop their search engines and indexes. As suggested, we
> > might
> > > try to patch things up with Rootsweb, or form a closer alliance with
> > (choose
> > > one... Ancestry, FamilySearch, FTK, KK, GEDSEARCH, et. al.) but regardless,
> > > we will not have autonomy or control of the direction they take.
> > >
> > > I believe that option # 2 can only occur after we have taken the steps to
> > > establish ourselves as a "real" entity. Regardless of the nature of that
> > > entity, The USGenWeb Project is not a legal entity that can negotiate any
> > > kind of agreement at this point.
> > >
> > > My idea in laying out my proposal was not to create a company that excluded
> > > the current membership, but as a means of creating something that we could
> > > all say "this is ours." We have built it, and we as a group benefit
> > from its
> > > growth and success.
> > >
> > > You know, one of the original concepts of this project was that the sites
> > > would be spread around on any number of servers, thus avoiding the
> > situation
> > > that has now developed where Rootsweb has had to build a mammoth site to
> > > handle the traffic. As conceived and essentially implemented, The USGenWeb
> > > Project could be continued to be operated essentially "free" on all of the
> > > "personal home page" sites each volunteer has, if we did not have 1/3
> > or more
> > > of the county pages (and all the archives) on the same server. My
> > Civil War
> > > in Georgia and Civil War in South Carolina sites are more than 100
> > megabytes
> > > each in size with each having something on the order of 3,000 pages,
> > and yet
> > > I have never had the least inkling from my provider in over 2 years now
> > that
> > > they are unhappy with the $25 per month I'm paying them.
> > >
> > > Would it be possible to implement a similar arrangement to the county pages
> > > with an index? There are something on the order of 12,000 viable soundex
> > > codes. Perhaps a procedure could be developed to pool and distribute the
> > > names down to a set of sites or servers so that the hits on the site
> > are not
> > > excessive, and we would avoid the costs of maintaining a high bandwidth
> > > search server site.
> > >
> > > As a further thought, the THUNDERSTONE local search engine I use on my
> > sites
> > > is totally free as long as you do not need to have them index more than 100
> > > megabytes or span multiple servers. We could create a set of top level
> > pages
> > > with the various surnames / soundex codes, or even a top level search
> > engine,
> > > which then cascades out to other sites much as the state pages do. So just
> > > as someone goes from national page, state page, county page now, they could
> > > say go from soundex code, to surname page (or search engine) showing the
> > > various spellings, and then to the links page (or search engine)
> > showing the
> > > links to the actual sites.
> > >
> > > What do you think? Is this a possibility?
> > >
> > > John Rigdon
>

This thread:

• [USGENWEB-SW] Re: Fwd: Re: USGW-CC-L: Search Engine Proposal Followup by "David W. Morgan" <>