Back before the search engines, the internet was relatively small but still growing enough that it needed searching. Computers were slower so the speed with which we expect results from Google was impossible. In order to search the internet, Gopher users had Archie, Veroncia and Jughead as well as lists of known Gopher servers – which linked to more lists of known Gopher servers. If you are interested in this system, it’s still online and a good start to look is http://gopher.floodgap.com/gopher/. Another way to search, however, was through Knowbots. These consist of several components
- A server (the “Knowbot Operating System”, or KOS) that runs on a host to enable it to run Knowbot programs. (In our terminology, this makes it a “Knowbot service station”.)
- A distributed, replicated namespace server (the “worldroot server”) which provides the shared namespace used by Knowbot programs for navigation and environmental inquiries.
- Tools to manage, create, submit, monitor and debug Knowbot programs.
- Tools to monitor and control collections of Knowbot service stations.
- A library of Python modules for use by Knowbot programs as well as by the support tools.
Usually to access a knowbot you telnet to a certain port and issue commands (and wait) or you email them and wait for a response.
The first knowbots up were Knowbot Information Services (KIS) to search for people.
The Knowbot Information Service (KIS) is another “white pages” service that performs a broad name search, checking MCI Mail, the X.500 White Pages Pilot Project, various Whois servers at various organizations (Whois is yet another directory service), and the UNIX finger command. It can be used either as a client program resident on your local machine, through e-mail, or by Telneting to a public server.
KIS uses subprograms called Knowbots to search for information. Each Knowbot looks for specific information from a site and reports back to the main program with the results.
Two hosts running KIS servers are info.cnri.reston.va.us and regulus.cs.bucknell.edu. You can access either one by electronic mail (send mail to netaddress@nri.reston.va.us, for instance) or using Telnet. (If you Telnet to a KIS server, you need to request port 185: instead of typing telnet regulus.cs.buckness.edu, you’d actually type telnet regulus.cs.buckness.edu 185.)
Because searching can take several minutes, I prefer to use the e-mail method; once KIS knows the results of the search, it mails them back to you.
In the body of your mail message to netaddress, put names of your associates, one per line. You may use first and last names or a login if you know them. Sending johnson will search the default list of directory servers for user johnson. Because KIS checks a predefined set of services, you do not need to supply an organization name to check for.
KIS also includes commands for narrowing your search and searching for an organization. For more help, include the word man in your e-mail to KIS or your interactive session.
Source: https://www.savetz.com/yic/YIC04FI_23.html
The University of Illinois had the following knowbot:
INTERNET ADDRESSES:
nri.reston.va.us 185
132.151.1.1 185
sol.bucknell.edu 185
134.82.1.8 185
DESCRIPTION:
Knowbot is an useful information service for locating
someone with an Internet address. Knowbot does not
have its own "white pages" recording internet users like a
telephone book. However, Knowbot can access to other
information services that have their own "white pages"
and search for you. Commands to operate knowbot service
are easy but not very user friendly to first time users.
SERVICES:
Knowbot serves as a gateway for internet users in remote hosts by
sending searching commands to find someone in internet, receiving the
searching results and presenting results in a uniform format for the
user. However, very often the Knowbot search is fruitless, because
of the incomplete information of internet users.
Listed below are remote host accessible to Knowbot. They all have
their own users information pools.
nic
mcimail
ripe
x500
finger
nwhois
mitwp
quipu-country
quipu-org
ibm-whois
site-contacts
LOGIN SEQUENCE:
At system prompt, type telnet nri.reston.va.us 185
systemprompt> telnet nri.reston.va.us 185
EXIT SEQUENCE:
To exit Knowbot, type RquitS at the Knowbot prompt.
>quit
ACCESS COMMANDS:
To enact command, type the command at Knowbot
prompt,
>[command]
e.g. >help
Access commands of Knowbot include:
>help
to print a summary of Knowbot commands on
screen
>man
to print an on-line manual of Knowbot on screen
>quit
to exit Knowbot information system
>[name]
to start searching a name of person with internet
address
e.g. >Krol
>services
to list all Knowbot accessible hosts
>service [hostname]
to narrow the search service on a specific host
e.g. > service nic
>org [organization]
to narrow the search service on a specific
organization
e.g. >org University of Illinois
>country [country name]
to narrow the search service on a specific country
e.g. >country US
SAMPLE LOGIN:
1. telnet to Knowbot at system prompt
systemprompt> telnet nri.reston.va.us 185
2. specify the organization of the person to be searched
> org university of Illinois
and/or you may specify the host service
> service nic
3. type in the name to start searching
> krol
4. You may get the following result:
Name: Ed Krol
Organization: University of Illinois
Address: Computing and Communications Service
Office,195 DCL, 1304 West Springfield
Avenue
City: Urbana
State: IL
Country: US
Zip: 61801-4399
Phone: (217) 333-7886
E-Mail: Krol@UXC.CSO.UIUC.EDU
Source: whois@nic.ddn.mil
Ident: EK10
Last Updated: 27-Nov-91
5. exit Knowbot
> quit
FRIENDLY ADVICE:
Since there are no complete recordings of all Internet
users, it is better not to expect to locate every internaut
through Knowbot. However, the more you know about
the person you want to locate, the easier the searching
process, because you can narrow the search by specifying
organization, country, or host of the person to be
searched, which will save you a lot of time.
DOCUMENT AUTHORS: Hsien Hu
Irma Garza
Source: https://www.ou.edu/research/electron/internet/knowbot.htm
These knowbots were developed before and during 1995 – NASA had plans for the Iliad knowbot (which gave me much better results than google, altavista, askjeeves or the other search engines of the time for specific tasks) back then.
https://ntrs.nasa.gov/api/citations/19970006511/downloads/19970006511.pdf
or https://www.linkielist.com/wp-content/uploads/2021/10/NASA-knowbots-iliad-19970006511-1.pdf
Iliad was developed as a resource for blind people but it waas realised that it worked well for teachers too. By sending an email to iliad@prime.jsc.nasa.gov you would receive the following reply:
about:blankPreformatted: Change block type or styleAdd titleKnowbots – the first way to search, relevant again?
Back before the search engines, the internet was relatively small but still growing enough that it needed searching. Computers were slower so the speed with which we expect results from Google was impossible. In order to search the internet, Gopher users had Archie, Veroncia and Jughead as well as lists of known Gopher servers – which linked to more lists of known Gopher servers. If you are interested in this system, it’s still online and a good start to look is http://gopher.floodgap.com/gopher/. Another way to search, however, was through Knowbots. These consist of several components
- A server (the “Knowbot Operating System”, or KOS) that runs on a host to enable it to run Knowbot programs. (In our terminology, this makes it a “Knowbot service station”.)
- A distributed, replicated namespace server (the “worldroot server”) which provides the shared namespace used by Knowbot programs for navigation and environmental inquiries.
- Tools to manage, create, submit, monitor and debug Knowbot programs.
- Tools to monitor and control collections of Knowbot service stations.
- A library of Python modules for use by Knowbot programs as well as by the support tools.
Usually to access a knowbot you telnet to a certain port and issue commands (and wait) or you email them and wait for a response.
The first knowbots up were Knowbot Information Services (KIS) to search for people.
The Knowbot Information Service (KIS) is another “white pages” service that performs a broad name search, checking MCI Mail, the X.500 White Pages Pilot Project, various Whois servers at various organizations (Whois is yet another directory service), and the UNIX finger command. It can be used either as a client program resident on your local machine, through e-mail, or by Telneting to a public server.
KIS uses subprograms called Knowbots to search for information. Each Knowbot looks for specific information from a site and reports back to the main program with the results.
Two hosts running KIS servers are info.cnri.reston.va.us and regulus.cs.bucknell.edu. You can access either one by electronic mail (send mail to netaddress@nri.reston.va.us, for instance) or using Telnet. (If you Telnet to a KIS server, you need to request port 185: instead of typing telnet regulus.cs.buckness.edu, you’d actually type telnet regulus.cs.buckness.edu 185.)
Because searching can take several minutes, I prefer to use the e-mail method; once KIS knows the results of the search, it mails them back to you.
In the body of your mail message to netaddress, put names of your associates, one per line. You may use first and last names or a login if you know them. Sending johnson will search the default list of directory servers for user johnson. Because KIS checks a predefined set of services, you do not need to supply an organization name to check for.
KIS also includes commands for narrowing your search and searching for an organization. For more help, include the word man in your e-mail to KIS or your interactive session.Source: https://www.savetz.com/yic/YIC04FI_23.html
The University of Illinois had the following knowbot:
INTERNET ADDRESSES:
nri.reston.va.us 185
132.151.1.1 185
sol.bucknell.edu 185
134.82.1.8 185
DESCRIPTION:
Knowbot is an useful information service for locating
someone with an Internet address. Knowbot does not
have its own "white pages" recording internet users like a
telephone book. However, Knowbot can access to other
information services that have their own "white pages"
and search for you. Commands to operate knowbot service
are easy but not very user friendly to first time users.
SERVICES:
Knowbot serves as a gateway for internet users in remote hosts by
sending searching commands to find someone in internet, receiving the
searching results and presenting results in a uniform format for the
user. However, very often the Knowbot search is fruitless, because
of the incomplete information of internet users.
Listed below are remote host accessible to Knowbot. They all have
their own users information pools.
nic
mcimail
ripe
x500
finger
nwhois
mitwp
quipu-country
quipu-org
ibm-whois
site-contacts
LOGIN SEQUENCE:
At system prompt, type telnet nri.reston.va.us 185
systemprompt> telnet nri.reston.va.us 185
EXIT SEQUENCE:
To exit Knowbot, type RquitS at the Knowbot prompt.
>quit
ACCESS COMMANDS:
To enact command, type the command at Knowbot
prompt,
>[command]
e.g. >help
Access commands of Knowbot include:
>help
to print a summary of Knowbot commands on
screen
>man
to print an on-line manual of Knowbot on screen
>quit
to exit Knowbot information system
>[name]
to start searching a name of person with internet
address
e.g. >Krol
>services
to list all Knowbot accessible hosts
>service [hostname]
to narrow the search service on a specific host
e.g. > service nic
>org [organization]
to narrow the search service on a specific
organization
e.g. >org University of Illinois
>country [country name]
to narrow the search service on a specific country
e.g. >country US
SAMPLE LOGIN:
1. telnet to Knowbot at system prompt
systemprompt> telnet nri.reston.va.us 185
2. specify the organization of the person to be searched
> org university of Illinois
and/or you may specify the host service
> service nic
3. type in the name to start searching
> krol
4. You may get the following result:
Name: Ed Krol
Organization: University of Illinois
Address: Computing and Communications Service
Office,195 DCL, 1304 West Springfield
Avenue
City: Urbana
State: IL
Country: US
Zip: 61801-4399
Phone: (217) 333-7886
E-Mail: Krol@UXC.CSO.UIUC.EDU
Source: whois@nic.ddn.mil
Ident: EK10
Last Updated: 27-Nov-91
5. exit Knowbot
> quit
FRIENDLY ADVICE:
Since there are no complete recordings of all Internet
users, it is better not to expect to locate every internaut
through Knowbot. However, the more you know about
the person you want to locate, the easier the searching
process, because you can narrow the search by specifying
organization, country, or host of the person to be
searched, which will save you a lot of time.
DOCUMENT AUTHORS: Hsien Hu
Irma Garza
Source: https://www.ou.edu/research/electron/internet/knowbot.htm
These knowbots were developed before and during 1995 – NASA had plans for the Iliad knowbot (which gave me much better results than google, altavista, askjeeves or the other search engines of the time for specific tasks) back then.
Information Infrastructure Technology Applications (IITA) Program Annual K-12 Workshop April 11 – 13 1995 (PDF)
Iliad was developed as a resource for blind people but it waas realised that it worked well for teachers too. By sending an email to iliad@prime.jsc.nasa.gov you would receive the following reply:
Your question has been received and is being processed by the ILIAD
knowbot.
Responses will be sent to the email address provided in the heading.
You can now specify
*outputtype: dwl
(document with links) to receive documents with embedded hot links in the
documents.
For example:
Subject: iliad query
*outputtype: dwl
?q: nasa jsc ltp
An example query response would consist of the documents found and a summary. It was surpisingly well curated. Here is an example summary:
Dear ILIAD User:
This is a summary of the documents sent to you by ILIAD in response to
your email question. The number order of the summarized documents
corresponds to the number on the individual documents you received.
Your question was:
internet bots automated retrieval=20
Output Type: documents
1)
"http://navigation.us.realnames.com/resolver.dll?action=resolution&charset
=utf-8&realname=TEKTRAN+%3A+USDA+Technology+Transfer+Automated+Retrieval+S
ystem&providerid=154"
TEKTRAN : USDA Technology Transfer Automated Retrieval
System TEKTRAN : USDA Technology Transfer Automated
Retrieval System: TEKTRAN : USDA Technology Transfer
Automated Retrieval System Click on this
Internet Keyword to go directly to the TEKTRAN : USDA
Technology Transfer Automated Retrieval System Web
site. 1000,http://www.nal.usda.gov/ttic/tektran/tektran.html
( Internet Keyword).+\( (\S+).*\) OCLC
Internet Cataloging Project Colloquium Field Report By
Amanda Xu MIT Libraries When we joined the OCLC Intercat Project, our
first concern was the feasibility of using MARC formats and AACR2 for
describing and accessing Internet resources of various
types. 999,http://www.oclc.org/oclc/man/colloq/xu.htm (
WebCrawler)
2) "http://www.botspot.com/faqs/article3.htm" BotSpot ® : The Spot
for all Bots & Intelligent Agents search botspot free
newsletter internet.com internet.commerce PAGE 3 OF
6 Beyond Browsing... Offline Web Agents by Joel T. Patz is an
excellent
article comparing the current Offline Web Agents and giving detailed
explanations and instructions including head-to-head feature
charts and downloading
sites. 888,http://www.botspot.com/faqs/article3.htm (
WebCrawler)
3) "http://www.insead.fr/CALT/Encyclopedia/ComputerSciences/Agents/"
Agent Technologies Agent
technologies
789,http://www.insead.fr/CALT/Ency...pedia/ComputerSciences/Agents/
( WebCrawler)
4) "http://lonestar.texas.net/disclaimers/aup.html" Acceptable
Use Policy Texas.Net Acceptable Use Policy In order for Texas
Networking to keep your service the best it can be, we have a set of
guidelines known as our "Acceptable Use Policy." These guidelines
apply to all customers equally and covers dialup account usage as well
as mail, news, and other
services. 480,http://lonestar.texas.net/disclaimers/aup.html
( WebCrawler)
5)
"http://navigation.us.realnames.com/resolver.dll?action=resolution&charset
=utf-8&realname=Automated+Traveller%27s+Internet+Site&providerid=154"
Automated Traveller's Internet Site
Automated Traveller's Internet Site: The
Automated Traveller-Discounted Airfares
Worldwide Click on this Internet Keyword to go directly to
the Automated Traveller's Internet Site Web
site. 333,http://www.theautomatedtraveller.com/ ( Internet
Keyword).+\( (\S+).*\) This site provides you with an
assortment of search devices along with their brief descriptions.
Also, you will find recommendations for using specific research tools
and their combinations that we have found more productive in our own
research. 284,http://www.brint.com/Sites.htm ( WebCrawler)
The following references were not verified for uniqueness.
You can retrieve any these references by sending ILIAD an email
request in the following format:
Subject: get url
url: <the url name>
for example:
Subject: get url
url: http://prime.jsc.nasa.gov/iliad/index.html
If you want embedded hot links in the document add "*outputtype: dwl"
before the first url: line
for example:
Subject: get url
*outputtype: dwl
url: http://prime.jsc.nasa.gov/index.html
1) "http://gsd.mit.edu/~history/search/engine/history.html" A
History of Search Engines What's a Robot got to do with the
Internet ? Other types of robots on the Internet push
the interpretation of the automated task definition. The
chatterbot variety is a perfect
example. 681,http://gsd.mit.edu/~history/search/engine/history.html
( WebCrawler)
2)
"http://navigation.us.realnames.com/resolver.dll?action=resolution&charset
=utf-8&realname=Automated+Information+Retrieval+Systems+%28AIRS%29&provide
rid=154"
Automated Information Retrieval Systems (AIRS)
Automated Information Retrieval Systems (AIRS):
Automated Information Retrieval Systems (AIRS) Click on
this Internet Keyword to go directly to the Automated
Information Retrieval Systems (AIRS) Web
site. 666,http://www.re-airs.com/ ( Internet Keyword)
.+\( (\S+).*\) The Internet Communications
LanguageTM News Events Technology 30-October-1999: Linux World A
REBOL Incursion It's not a scripting language, not a programming
language -- and not a new Amiga,
either. 584,http://www.rebol.com/inthenews.html (
WebCrawler)
3)
"http://www.pcai.com/pcai/New_Home_Page/ai_info/intelligent_agents.html"
PC AI - Intelligent Agents Requires Netscape 2.0 or later
compatibility. Intelligent Agents execute tasks on behalf of a
business process, computer application, or an
individual.
384,http://www.pcai.com/pcai/New_H...i_info/intelligent_agents.html
( WebCrawler)
4) "http://www.rci.rutgers.edu/~brcoll/search_engines.htm"
Searching with Style Motto for the Day: Hypberbole n:
extravagant exaggeration; see also computer industry. Last Updated:
November 10, 1996 Very few aspects of the Web are developing as fast
as
the search engines, except for the sheer volume of
information. 186,http://www.rci.rutgers.edu/~brcoll/search_engines.htm
( WebCrawler)
5) "http://www.aci.net/kalliste/echelon/ic2000.htm" STOA Report:
Interception Capabilities 2000 Interception Capabilities
2000 Report to the Director General for Research of the European
Parliament (Scientific and Technical Options Assessment programme
office) on the development of surveillance technology and risk of
abuse of economic
information. 89,http://www.aci.net/kalliste/echelon/ic2000.htm
( WebCrawler)
Thank you for using ILIAD. This marks the end of your results.
5 files passed analysis.
Search performed by metacrawler.
End of ILIAD Session ID: SEN38899
---------------------------------------------------------
Illiad could be searched through tenet and msstate and a few other providers:
You can use the well-known e-mail meta-finder
ILIAD (Internet Library Information Access Device) knowbot,
which can be found at <iliad@msstate.edu> or
<iliad@algol.jsc.nasa.gov>. You will receive instructions at the request of
“startiliad” in the subject of the message.
The query sent to the ILIAD server will be sent to several largest
search servers (eg Altavista, Excite, InfoSeek, Lycos,
WebCrawler, …) removes duplicate and overly
irrelevant documents from the results , and
sends the already downloaded pages (without graphics) back within 15 – 20 minutes. You can also try ILIAD on the WWW,
via the form at
<http://www.tenet.edu/library/iliad.html>.
A list of email services can be found here but is copied as these pages are going down pretty quickly
Get webpages via eMail
Several years ago when the Internet connections where slow and the “www” just invented, many people just got a to email restricted access to the Internet. That’s the origin of the “Agora” and “www4email” software. Some of these email robots are still available and we can use them to bypass Internet censorship. The best thing would be to subscribe to a free email provider which allows SSL-connections (like https://www.fastmail.fm/, https://www.ziplip.com/, https://www.hushmail.com/, https://www.safe-mail.net/, https://www.mail2world.com/, https://www.webmails.com/ e.t.c) and use that account with the email addresses below. I put the field where you have to input the URL in brackets. It still works great for text. But sure there are big problems with images or even DHTML, JavaScript, Java, Flash e.t.c. Also other services besides www are possible, for a very good tutorial on this see ftp://rtfm.mit.edu/pub/usenet/news.answers/internet-services/access-via-email. There is also a web based service under http://www.web2mail.com/. I again used www.web.freerk.com/c/ as an example because the URL is all time accessible and the ‘.com’ in the original Google address is often considered as a .com DOS-file by some computers and censorship systems. The www4mail software (http://www.www4mail.org/) is newer than the Agora software.
A eMail with just “help” in the subject line will get you a tutorial on howto use the service properly.
page@grabpage.org
[SUBJECT] url: http://www.web.freerk.com/c/
info: http://www.grabpage.org/
frames@pagegetter.com
[BODY] http://www.web.freerk.com/c/
info: http://www.pagegetter.com/
web@pagegetter.com
[BODY] http://www.web.freerk.com/c/
info: http://www.pagegetter.com/
webgate@vancouver-webpages.com
[BODY] get http://www.web.freerk.com/c/
info: http://vancouver-webpages.com/webgate/
webgate@vancouver-webpages.com
[BODY] mail http://www.web.freerk.com/c/
info: http://vancouver-webpages.com/webgate/
www4mail@wm.ictp.trieste.it
[BODY] http://www.web.freerk.com/c/
info: http://www.ictp.trieste.it/~www4mail/
www4mail@access.bellanet.org
[BODY] http://www.web.freerk.com/c/
info: http://www.bellanet.org/email.html
www4mail@kabissa.org
[BODY] http://www.web.freerk.com/c/
info: http://www.kabissa.org/members/www4mail/
www4mail@ftp.uni-stuttgart.de
[BODY] http://www.web.freerk.com/c/
www4mail@collaborium.org
[BODY] http://www.web.freerk.com/c/
info: http://www.collaborium.org/~www4mail/
binky@junoaccmail.org
[BODY] url http://www.web.freerk.com/c/
info: http://boas.anthro.mnsu.edu/
iliad@prime.jsc.nasa.gov
[SUBJECT] GET URL
[BODY] url:http://www.web.freerk.com/c/
info: http://prime.jsc.nasa.gov/iliad/
Google Search via eMail:
google@capeclear.com
[Subject] search keywords
info: http://www.capeclear.com/google/
More info: http://www.cix.co.uk/~net-services/mrcool/stats.htm
ftp://rtfm.mit.edu/pub/usenet/news.answers/internet-services/access-via-email
Information by Fravia on building them can be found https://www.theoryforce.com/fravia/searchlores/bots.htm – there seems to be a copy up to phase five at http://www.woodmann.com/fravia/botstart.htm
A complete knowbot software suite can be downloaded from https://www.cnri.reston.va.us/home/koe/index.html. This was written by the CNRI [1].
Knowbot programming: System support for mobile agents is another useful overview
A short history (in Czech) can be found here: Vše, co jste chtěli vědět o Internetu… nebojte se zeptat!
Today with the volume of information on the web being so huge, there may be a market for a resurgence of this kind of software. Google realises that it’s fast become impossible to find what you are looking for accurately and has responded by having specific search engines (eg scholar, books, images, shopping, etc) for specific tasks. However for specific fields this is still way too large. A way to handle this would be to have semi-curated search sources added to a knowbot within a very specific field (eg energy, psychology, hardware) allowing you to search easily within expertise. If you can then heuristically detect which field is being searched you can direct the searcher to that specific knowbot.
