On Sat, 18 May 2013, Ian McLaughlin wrote:
(I may be too late answering this, because I'm mobile and there's a billion messages in this thread that I haven't read yet...)
I'd really like to see a Hecnet-based solution - like a set host to a captured account on mim or something would be nice. A web page kind of breaks the illusion of an antique network :). Limiting access to hecnet methods sort of self-polices just by itself.
A "captured" account? ;)
Ian
(Trying to keep up with the tread while at my son's birthday party)
Sent from my iPhone
On 2013-05-18, at 9:58 AM, Johnny Billquist <bqt at softjar.se> wrote:
On 2013-05-18 18:52, Cory Smelosky wrote:
On Sat, 18 May 2013, Dave McGuire wrote:
On 05/18/2013 12:45 PM, Johnny Billquist wrote:
I like the idea of giving us all write access. Maybe you can set up
regular dumps of the database for restoration in case someone screws
up.
I can certainly do that. It just becomes a question of people might
need to
reenter information in case a rollback has to go far. Also trickier to
realize if "corruption" actually have happened perhaps.
On the other hand, I could atleast have permissions that only allowed
random
people to modify data, not add new, nor delete.
That's a really good idea.
I second this. How flexible is Datatrieve's permissions system?
You can identify users either by UIC, or by entity-specific passwords. And for each type of thing in there you have Read, Write, Modify, Extend, and Control. And the access is a list, with the first match being the permissions used.
So I could either make sure all people who might want to modify this have their own account on MIM, or else I could just create password protected access, and we could have shared or private passwords.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
---
Filter service subscribers can train this email as spam or not-spam here: http://my.email-as.net/spamham/cgi-bin/learn.pl?messageid=2C48D1D4BFDC11E2A…
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
(I may be too late answering this, because I'm mobile and there's a billion messages in this thread that I haven't read yet...)
I'd really like to see a Hecnet-based solution - like a set host to a captured account on mim or something would be nice. A web page kind of breaks the illusion of an antique network :). Limiting access to hecnet methods sort of self-polices just by itself.
Ian
(Trying to keep up with the tread while at my son's birthday party)
Sent from my iPhone
On 2013-05-18, at 9:58 AM, Johnny Billquist <bqt at softjar.se> wrote:
On 2013-05-18 18:52, Cory Smelosky wrote:
On Sat, 18 May 2013, Dave McGuire wrote:
On 05/18/2013 12:45 PM, Johnny Billquist wrote:
I like the idea of giving us all write access. Maybe you can set up
regular dumps of the database for restoration in case someone screws
up.
I can certainly do that. It just becomes a question of people might
need to
reenter information in case a rollback has to go far. Also trickier to
realize if "corruption" actually have happened perhaps.
On the other hand, I could atleast have permissions that only allowed
random
people to modify data, not add new, nor delete.
That's a really good idea.
I second this. How flexible is Datatrieve's permissions system?
You can identify users either by UIC, or by entity-specific passwords. And for each type of thing in there you have Read, Write, Modify, Extend, and Control. And the access is a list, with the first match being the permissions used.
So I could either make sure all people who might want to modify this have their own account on MIM, or else I could just create password protected access, and we could have shared or private passwords.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
---
Filter service subscribers can train this email as spam or not-spam here: http://my.email-as.net/spamham/cgi-bin/learn.pl?messageid=2C48D1D4BFDC11E2A…
On Sat, 18 May 2013, Paul_Koning wrote:
On May 18, 2013, at 1:28 PM, Cory Smelosky wrote:
On Sat, 18 May 2013, G. wrote:
...
If you use the CTERM protocol, it will be always slow and sluggish, no matter
how much you'll try to tweak it. Try SET HOST /APP=R and appreciate speed :)
It's MUCH more usable when using RTERM. Thank you.
Why is CTERM so slow on TOPS-10?
Because CTERM was designed, in essence, as a remote procedure call representation of the VMS terminal driver features. The ones in TOPS-10 are entirely different -- just consider the way carriage return echoes. So TOPS-10 is forced to do its terminal I/O through a protocol that doesn't come close to matching what it really wants to see. I suspect it simply runs things in single character mode, since that's about the only way to guarantee that you get what you want.
Ahhhh. Yeah. That would explain it.
Come to think of it, TOPS-10 RTERM mode may well be just that, but it's a very lightweight protocol so it's fast. It also has the nice benefit that it exists on systems that don't do CTERM -- like RSTS.
Yup. Very beneficial on systems without CTERM. ;)
paul
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On Sat, 18 May 2013, Cory Smelosky wrote:
On Sat, 18 May 2013, Dave McGuire wrote:
On 05/18/2013 01:49 PM, Cory Smelosky wrote:
On Sat, 18 May 2013, h vlems wrote:
Unfortunately not Cory, 0.22 euro per kWhr.
No way I can leave even one (real) system on line all day long.
That's more expensive than here, surprisingly.
What are you paying over there? It's $0.095/kWh here, was $0.135/kWh
down in Florida. (that was a major factor in my decision to move up here)
They don't give a proper breakdown of costs per kwh...but my (likely
incorrect) calculations showed it to be about $0.12/kwh.
55kwh/day at $7.02/day. Due to some odd plan account budget thing, it
amounts to much less than that in reality. Save for the month or so where
you make up the difference.
There's also this line on the bill:
"In order to save you money, a new supplier must offer you a price lower than 7.6 cents per KWH for the same usage that appears on this bill"
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On May 18, 2013, at 1:28 PM, Cory Smelosky wrote:
On Sat, 18 May 2013, G. wrote:
...
If you use the CTERM protocol, it will be always slow and sluggish, no matter
how much you'll try to tweak it. Try SET HOST /APP=R and appreciate speed :)
It's MUCH more usable when using RTERM. Thank you.
Why is CTERM so slow on TOPS-10?
Because CTERM was designed, in essence, as a remote procedure call representation of the VMS terminal driver features. The ones in TOPS-10 are entirely different -- just consider the way carriage return echoes. So TOPS-10 is forced to do its terminal I/O through a protocol that doesn't come close to matching what it really wants to see. I suspect it simply runs things in single character mode, since that's about the only way to guarantee that you get what you want.
Come to think of it, TOPS-10 RTERM mode may well be just that, but it's a very lightweight protocol so it's fast. It also has the nice benefit that it exists on systems that don't do CTERM -- like RSTS.
paul
On Sat, 18 May 2013, Dave McGuire wrote:
On 05/18/2013 01:49 PM, Cory Smelosky wrote:
On Sat, 18 May 2013, h vlems wrote:
Unfortunately not Cory, 0.22 euro per kWhr.
No way I can leave even one (real) system on line all day long.
That's more expensive than here, surprisingly.
What are you paying over there? It's $0.095/kWh here, was $0.135/kWh
down in Florida. (that was a major factor in my decision to move up here)
They don't give a proper breakdown of costs per kwh...but my (likely incorrect) calculations showed it to be about $0.12/kwh.
55kwh/day at $7.02/day. Due to some odd plan account budget thing, it amounts to much less than that in reality. Save for the month or so where you make up the difference.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On 05/18/2013 01:49 PM, Cory Smelosky wrote:
On Sat, 18 May 2013, h vlems wrote:
Unfortunately not Cory, 0.22 euro per kWhr.
No way I can leave even one (real) system on line all day long.
That's more expensive than here, surprisingly.
What are you paying over there? It's $0.095/kWh here, was $0.135/kWh down in Florida. (that was a major factor in my decision to move up here)
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
On Sat, 18 May 2013, G. wrote:
On Sat, 18 May 2013 06:14:27 -0000, you wrote:
SET HOST to MARLEY (9.10) (It's a bit slow to respond only via DECnet. I
might need to play with buffers)
If you use the CTERM protocol, it will be always slow and sluggish, no matter
how much you'll try to tweak it. Try SET HOST /APP=R and appreciate speed :)
It's MUCH more usable when using RTERM. Thank you.
Why is CTERM so slow on TOPS-10?
Credentials are guest and the password is guest. Still need to install some
stuff for language support.
If you like it more, there is a way to have a user with no password (and/or
with the ability to actually change password disabled). Just play with REACT
and look for the right incantation, because I do not remember it anymore...
It's related to requirements I believe.
G. :)
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments