Mailing List Archive

On Wed, 18 Oct 1995, Ben Laurie wrote:

> Does anyone know where I can get wu-ftpd? I've tried wuarchive.wustl.edu
> but I can't find it. Web searchers don't seem to be helping much either.

??? that's odd. I will try to find and respond to you personally.

> BTW, anyone tried this Hyper-G gadget?

I tried this from Berlin DE while I was over there first
half of 1995. The upside is that "it's all in the server".
The downside is "it's all in the server". I like the
user interface paradigm in their (one-and-only I think)
browser, but all aspects of the browser have to be Hyper-G
"aware" - that is, no such thing as "helper apps" which are
independent of the xaction protocols. :-( I think this work
grew up concurrently with the WWW, and they are hip enough
to recognize the success of the Web and provide a Web
interface (you dont get full functionality, but you get
SOMETHING). BTW, I think a number of Euro professional organizations
are using it as their archive handler (i.e. it has found users)

... oops, lost my URL for it. I can try to find and post
if anyone is interested.

dave
+------------------------------------------------------------------+
| David Oliver dave@newshare.com |
| Managing Director-Technology Newshare Corporation |
+------------------------------------------------------------------+

>
> BTW, anyone tried this Hyper-G gadget?

NCSA is doing collaborative work with the Hyper-G people at Gratz University
in Germany. What do you want to know?


From David Oliver:
> I tried this from Berlin DE while I was over there first
> half of 1995. The upside is that "it's all in the server".
> The downside is "it's all in the server". I like the
> user interface paradigm in their (one-and-only I think)
> browser, but all aspects of the browser have to be Hyper-G
> "aware" - that is, no such thing as "helper apps" which are
> independent of the xaction protocols. :-( I think this work
> grew up concurrently with the WWW, and they are hip enough
> to recognize the success of the Web and provide a Web
> interface (you dont get full functionality, but you get
> SOMETHING). BTW, I think a number of Euro professional organizations
> are using it as their archive handler (i.e. it has found users)
>
> ... oops, lost my URL for it. I can try to find and post
> if anyone is interested.


Their URL is : http://hyperg.iicm.tu-graz.ac.at/Chyperg

(Note: I tried to check the URL since I haven't been to the Hyper-G
site in a while, but the download was slow enough I had to leave for
a meeting before it finished. My apologies if the URL is no longer
valid.)

They have a lot of nice features. Their biggest problem lies with
scalability. Since every server knows about all the documents in every
server it is easy to maintain consistency, and provide link maps, but
it just doesn't scale well. The client uses a session manager which
brings up windows for each service requested. You can link a jpeg,
audio track, and document together to be a single object. They do
content negotiation on at least the language field and some other
fields. We've modified some of their code for use with some of our
client applications so we both can embed links in video now. I think
some of their viewers are being modified to work with Mosaic. To
reach WWW a Hyper-G client goes through an HTTP gateway. The Repository
project has been trying to include several of their ideas.

--
Elizabeth(Beth) Frank
NCSA Server Development Team
efrank@ncsa.uiuc.edu

> They have a lot of nice features. Their biggest problem lies with
> scalability. Since every server knows about all the documents in every
> server it is easy to maintain consistency, and provide link maps, but
> it just doesn't scale well.

Yep, that matches my evaluation as well. Some of the basic technology
is good (they learned a lot from the early Web mistakes) but they
fail to grasp the concept of scale (and, to a lesser extent, extensibility).
Aside from being link-server based, the Hyper-G project has been
centered almost exclusively within Graz and does not have the same
tradition of openness.

However, their biggest problem is their refusal to accept that
the Web is bigger than Hyper-G can handle. I think Hyper-G would make
a great software development environment or organizational repository,
but the link-server design prevents the robustness and complete
distributability of the Web. They could also make it a good server
management environment, were it not for their insanely stupid idea
of including client state in every URL sent by their gateway.

Note also that these people are behind the "Web Society" thingy
(along with the Gopher folks).

.....Roy

>
> >
> > BTW, anyone tried this Hyper-G gadget?
>
> NCSA is doing collaborative work with the Hyper-G people at Gratz University
> in Germany. What do you want to know?

I'd quite like to know a few things:

1. When are they going to fix their Web site? It doesn't work at all, apart
from the top page.

2. When are they going to work with proxies? I can't use Hyper-G at all without
that. I can't even download it.

3. When will source be available? (And if the answer is never they might as
well give up now, I reckon).

4. Why does everyone have an opinion on Hyper-G, and noone on wu-ftpd which
was my REAL question? :-)

>
>
> From David Oliver:
> > I tried this from Berlin DE while I was over there first
> > half of 1995. The upside is that "it's all in the server".
> > The downside is "it's all in the server". I like the
> > user interface paradigm in their (one-and-only I think)
> > browser, but all aspects of the browser have to be Hyper-G
> > "aware" - that is, no such thing as "helper apps" which are
> > independent of the xaction protocols. :-( I think this work
> > grew up concurrently with the WWW, and they are hip enough
> > to recognize the success of the Web and provide a Web
> > interface (you dont get full functionality, but you get
> > SOMETHING). BTW, I think a number of Euro professional organizations
> > are using it as their archive handler (i.e. it has found users)
> >
> > ... oops, lost my URL for it. I can try to find and post
> > if anyone is interested.
>
>
> Their URL is : http://hyperg.iicm.tu-graz.ac.at/Chyperg
>
> (Note: I tried to check the URL since I haven't been to the Hyper-G
> site in a while, but the download was slow enough I had to leave for
> a meeting before it finished. My apologies if the URL is no longer
> valid.)
>
> They have a lot of nice features. Their biggest problem lies with
> scalability. Since every server knows about all the documents in every
> server it is easy to maintain consistency, and provide link maps, but
> it just doesn't scale well.

"doesn't scale well" is understating, I think...

> The client uses a session manager which
> brings up windows for each service requested. You can link a jpeg,
> audio track, and document together to be a single object. They do
> content negotiation on at least the language field and some other
> fields. We've modified some of their code for use with some of our
> client applications so we both can embed links in video now. I think
> some of their viewers are being modified to work with Mosaic. To
> reach WWW a Hyper-G client goes through an HTTP gateway. The Repository
> project has been trying to include several of their ideas.

What's the Repository project (really showing my ignorance now :-)

Cheers,

Ben.

--
Ben Laurie Phone: +44 (181) 994 6435
Freelance Consultant Fax: +44 (181) 994 6472
and Technical Director Email: ben@algroup.co.uk
A.L. Digital Ltd,
London, England.

>
> >
> > BTW, anyone tried this Hyper-G gadget?
>
> NCSA is doing collaborative work with the Hyper-G people at Gratz University
> in Germany. What do you want to know?

I'd quite like to know a few things:

1. When are they going to fix their Web site? It doesn't work at all, apart
from the top page.

2. When are they going to work with proxies? I can't use Hyper-G at all without
that. I can't even download it.

3. When will source be available? (And if the answer is never they might as
well give up now, I reckon).

4. Why does everyone have an opinion on Hyper-G, and noone on wu-ftpd which
was my REAL question? :-)

>
>
> From David Oliver:
> > I tried this from Berlin DE while I was over there first
> > half of 1995. The upside is that "it's all in the server".
> > The downside is "it's all in the server". I like the
> > user interface paradigm in their (one-and-only I think)
> > browser, but all aspects of the browser have to be Hyper-G
> > "aware" - that is, no such thing as "helper apps" which are
> > independent of the xaction protocols. :-( I think this work
> > grew up concurrently with the WWW, and they are hip enough
> > to recognize the success of the Web and provide a Web
> > interface (you dont get full functionality, but you get
> > SOMETHING). BTW, I think a number of Euro professional organizations
> > are using it as their archive handler (i.e. it has found users)
> >
> > ... oops, lost my URL for it. I can try to find and post
> > if anyone is interested.
>
>
> Their URL is : http://hyperg.iicm.tu-graz.ac.at/Chyperg
>
> (Note: I tried to check the URL since I haven't been to the Hyper-G
> site in a while, but the download was slow enough I had to leave for
> a meeting before it finished. My apologies if the URL is no longer
> valid.)
>
> They have a lot of nice features. Their biggest problem lies with
> scalability. Since every server knows about all the documents in every
> server it is easy to maintain consistency, and provide link maps, but
> it just doesn't scale well.

"doesn't scale well" is understating, I think...

> The client uses a session manager which
> brings up windows for each service requested. You can link a jpeg,
> audio track, and document together to be a single object. They do
> content negotiation on at least the language field and some other
> fields. We've modified some of their code for use with some of our
> client applications so we both can embed links in video now. I think
> some of their viewers are being modified to work with Mosaic. To
> reach WWW a Hyper-G client goes through an HTTP gateway. The Repository
> project has been trying to include several of their ideas.

What's the Repository project (really showing my ignorance now :-)

Cheers,

Ben.

--
Ben Laurie Phone: +44 (181) 994 6435
Freelance Consultant Fax: +44 (181) 994 6472
and Technical Director Email: ben@algroup.co.uk
A.L. Digital Ltd,
London, England.

Ben Laurie writes:

> 4. Why does everyone have an opinion on Hyper-G, and noone on wu-ftpd which
> was my REAL question? :-)

From Lycos : <URL:http://freeway.oact.hq.nasa.gov/wuarchive.html>

--
Ryan L. Watkins email: vamp@dnx.com
HTML Hacker / System Admin url: http://www.dnx.com/vamp/vamp.html
Dimension X, Inc. phone: 415.296.0100

>
> Ben Laurie writes:
>
> > 4. Why does everyone have an opinion on Hyper-G, and noone on wu-ftpd which
> > was my REAL question? :-)
>
> From Lycos : <URL:http://freeway.oact.hq.nasa.gov/wuarchive.html>

Nice try. Like most other links I have obtained from Lycos, this one is yet
another mail archive, with a broken pointer to the distribution
(http://fegmania.wustl.edu/ftp.html). It hasn't been updated since April, by
the looks of things.

> Ryan L. Watkins email: vamp@dnx.com

Cheers,

Ben.

--
Ben Laurie Phone: +44 (181) 994 6435
Freelance Consultant Fax: +44 (181) 994 6472
and Technical Director Email: ben@algroup.co.uk
A.L. Digital Ltd,
London, England.