What is the usual transfer rate of FTP? Everytime I FTP a file, I only
get like .3K/sec or something? I heard of transfer rates of 200M/Sec
elsewhere...

61 responses total.

I expect it won't be any faster than the weakest link in the chain
(network) which in this case would probably be whatever portion
of Grex's 28.8 modem you're using at the moment.  I've timed ftp
transfers on my 2400 modem and my 14.4 modem using NCSA Telnet & a
PPP dialin connection.  They weren't any faster than a Zmodem transfer
over dialin (and often slower).  Now, ftp transfers over an ethernet
connection are *fast*...but then that's a faster link...

I get up to about 1300 bps with MacPPP to the NAS and a 14.4 3.2bis
4.2bis modem.

I've gotten up to 1600 bps with serial dialin and a 14.4 3.2/4.2 bis
modem and Zmodem.  Usually it's around 1200, however.  There seems
to be no rhyme or reason to why it's as slow as 120 bps one minute
and 1300 bps the next (during the same file download).

Well, actually, when you FTP some file, it copies the file onto Grex's
system, then I can download it to my system... what I meant was the 
speed of the NET... After I FTP some file, it says that it is like 
.3K per sec... I thinks that's pretty slow for Internet...

When you split 28.8 24 (or more) ways...(even with compression)...

You are assuming that our system is attached to the internet directly
via an ethernet card as is often the case in a university setting.
We are not. We are connected by a phone line to a commercial internet
provider. Lots of traffic is competing with your ftp packets for that 
narrow (28.8 no header compression) little pipe, so it will be very slow.

Our link may get enhanced some, but it will likely never attain the
speed of a university site. Grex just isn't a really good place to
exercise ftp, I'm afraid. You may see some improvement by choosing a
time when fewer people are on using the link, too.

Oh, I see... I thought Grex was connected directly... not by some 
provider...

These days, there really is no such thing as a "direct" connection
to "the Internet", since there really isn't any monolithic "the Internet"
to connect to.  The closest you can get is the ANS T3 backbone,
but even that isn't "the Internet" -- a large number of service
providers are connected to the CIX's backbone rather than ANS's backbone,
for instance.

For us, though, the limiting factor will almost certainly be our
connection to the rest of the net for the forseeable future.  Other
sites will almost always have a fast connection than we do.

I think that's pretty bad that Grex is slowed down like that... owell..

I wouldn't call it "bad". Grex has never intended to be an "internet
provider". It is a computer conferencing system - to which an internet
link for distant friends to use was added relatively recently. 

If you don't like the speed of Grex's Internet connection, the
most effective thing you can do is give us funding for a faster
link.  Right now the thing keeping us from getting a faster
link isn't political or technical, but financial -- we just can't
afford a T1 or T3 speed link (which can cost tens or hundreds of
thousands of dollars a year).

I think Grex's internet link has been performing wonderfully.
While a much faster link (is there anything in between our 28.8
and T1/T3 service?) would be even more wonderful, what we have now
is a far cry from a year ago...

I think it's possible to go with a 56k service using a leased line.
I believe there are local internet service providers offering that
service, but the cost of even that is hundreds of $/month.
It may also be possible to use ISDN to obtain roughly 128kb
service. I have no idea about cost and availability on that,
but I know it comes with a start-up cost for ISDN interface equipment.

well, I've been sending my donation in... But I doubt $6 a month is 
enough for a really big speed link... 

The common steps are:
        * modem-speed dialup links (up to 28.8Kbps)
        * 64Kbps leased line
        * 128Kbps ISDN
        * fractional T1 (64Kbps steps)
        * 1.5Mbps T1
        * fractional T3 (1.5Mbps steps)
        * 45Mbps T3
        * Gbps various fiber media (FDDI, ATM over SONET, etc.)
I doubt Grex will ever have anything faster than T1-speed.

I'd love to say "Don't worry, Grex will never need a T1", but I
know I'd regret it later.  >8)

So how much for the leased line?  The ISDN?

Yeah, how much is one of those leased lines or the ISDN?

As someone already mentioned...hundreds or thousands per month
depending...

A 64Kbps leased line is usually in the $100-$200/month range.
Two ISDN endpoints vary widely, price-wise, but count on
at least $30/month/endpoint.  Each of these also has high
capital costs, of course -- to use a leased line you need
a CSU/DSU at each end, and to use an ISDN line you need ISDN
codecs.  Codecs are $1000-$2000 each, and CSU/DSU's are up there, too.

(Anybody want to translate that?)

Read :18...

Well, $100-$200 is not bad...

But more than Grex can afford right now...

Well, if we get more people to donate ... maybe we CAN!!!

zlee, gotta get the hang of Grex here...they don't think like that.
They'd rather deal with day to day problems, and limit their thinking
by the funds available right now.  You're just too way out for this
system; you're making too much sense...

(zlee, if you are under the impression that Grex *ISN'T* getting donations
already, you would be *WELL ADVISED* to take a look at the coop
conference, which is where much of the discussion regarding the future of
Grex is taking place.)

(kentn, I take *SERIOUS* offense to your earlier statement. You make it
sound as if no one wants Grex to grow, and I *hope* that isn't what you
meant. if Grex was as bad as you implied, *no one* would be here, i.e.,
there wouldn't be a Grex. it would have gone the way of earlier BBSes.)

I would encourage anyone interested in the operation and planning of Grex
to read the coop conference. If you do, you'll find that very recently
there as been a surge in donations resulting in an accumulation of more
cash than we have seen in one place at one time since I can remember.

We still do not yet know whether this was a transient effect, although
I suspect it may not be. Caution is always advised. Improving the internet
connection and establishing a "rainy day" cash account are two of many
items related to our planning that are discussed online in coop.
In addition, these will be discussed in our monthly ftf board meeting
which is announced in the motd. 

I don't think #25 is fair either. We might make decisions faster if this
place were run by an autocrat, but it's not. It's run by you!
So the wheels grind slowly, but they do grind, and we are planning.

Notes for the acronym impaired:
ftf = face-to-face
motd = message of the day, what you see when you log in.

Guess what?  I don't give a flying f*** what you think about the
"fairness" of my statement!  There is a basis in Grex's history for
my interpretation...get yer act together and do some planning for
*growing* the system instead of riding the waves of success until
they crash on the beach.

(kentn, I *STRONGLY* feel that you are wrong on this point.)

(when there was a feeling of a need for more disk space, the call for
money was sent and _well-recieved_. because of this, there WILL be more
diskspace on-line, and shortly, from what I understand. this wasn't
because Grex had been counting pennies on its way to affording something
more; it was because we ASKED ourselves for donations, and those donations
came rolling in.)

(I also *SINCERELY* doubt that Grex wants to get bogged down in day-to-day
things, esp. staff, unless all that talk about disk quotas and
auto-reapers is in my head, of course. I know that I'd personally rather
look to the future, despite not being very fond of what has come so far. I
believe that most other Grexers are the same.)

(perhaps you would have been happier on the old M-Net. sure, things usu.
got done there, but look at them now. if you don't know, there are several
items on the history of M-Net in... the history conf.)

(Grex didn't happen solely on a whim.)

Hey, kentn, lay off a little... I know it takes a lot of planning 
to do something big... such as like getting a faster line or something...
We just have to hope that the Grex management is doing their part to 
make Grex better than it is now... (if it isn't already, just a little
slow)... I'm doing my part... that's all I have to say...

(zlee, we ARE the management. Grex is a cooperative.)

The only "unfairness" in #25 is that the statement "they don't think like that"
is too categorical. There are some that do think like that. However, some
also think that bigness for its own sake is not a worthwhile goal. Hence,
one needs a reason for getting bigger that is founded in the purposes
of the corporation. I think it would be fair to observe the *too much*
bigness could change the nature of Grex, as at some level, staff would
have to be hired, and finances would become such a big issue that members
would become less involved as administrators are hired. So - how big is
big enough?

For #31, ok... what do we need to do to make Grex better?

Kent, if you're disappointed with the level of planning that goes
on here, I suggest you do something about it.  Gather statistics,
do some planning for yourself, and present 1-year, 3-year, and 5-year
outlooks for Grex at the next board meeting.   Please include things
like estimated annual income, estimated annual costs, etc.  You can
include recommendations for new equipment to handle your projected
use rates, if you like, as well as a timetable for acquiring it
(keeping in mind the cost of the equipment vs. projected income,
of course).

I don't know about you, but I do Grex as a *hobby*.  As such, if it
ceases to be fun and enjoyable, I'll probably stop.  In addition,
because it is a hobby, I tend to do the parts that I enjoy and sort of
skip over the parts I dislike.  Drawing up long-term outlooks and
trying to predict how many people we'll have using the system 5 years
from now is not something I find enjoyable, so I don't do it.  (I also
don't think it's possible to do accurately -- no one had any idea how
the Internet connection would make Grex's size increase to the level it
did.)

It's not just 1/3/5 year plans that need to be done.  More basic
planning needs to be done...the kind of thinking and deciding that
generate that annual income and incur those annual costs.  As a
matter of fact, I *do* enjoy such planning and forecasting.  But
it is more than a one person job.

Yes, it requires information - and opinions - from all those that are
working on those parts that they enjoy. It is exactly *because* it is a
hobby, that collective and cooperative planning is required - no one
person knows all the facets of the operation well enough to concoct a
fully developed plan for practically anything. But, working together,
those separate facets of expertise can be assembled into useful plans.

#25 says we don't do any planning here. It's false.
I guess we'll have to disagree on this.
Could we plan more than we are doing? I'm sure we could.
I think we need to and I think we will.
I strongly recommend you come to a meeting and pitch in.
Some people think grex is run by an inner circle.
This is way wrong. We will appreciate the help of anyone
who can contribute. If there's any reason we haven't
spent enough time planning, it's because no one has had
the time to spend on it.

Grex *is* run by an inner circle - anyone willing to get involved &
work on it.  I for one don't think this is all that bad.  A more
restricted inner circle - with authority (or power) to *demand*
financial contributions and to implement plans whether or not they
please everyone else -  would likely do a lot more of the kind of
planning that Kent says we don't do - & Kent would be at the top of the
list of those complaining about it.  Kent, you can't have it both ways.

Sure I can.  Complaining doesn't require having any particular way.

If you would spend less time trying to find reasons why I'm wrong
and more time trying to see the truth in my statements, you and Grex
might get somewhere.  Grex needs to do more formal planning...
 

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss