Subject: Re2: Responses and suggestions on Raith reservation policy
From: James Conway <jwc@snf.stanford.edu>
Date: Wed, 01 Dec 2004 15:55:43 -0800
Wed, 01 Dec 2004 15:55:43 -0800
Greetings:

Interesting discussion on this topic thread. 
I feel the fewer rules the better, and resist all suggestions to add 
more rules as people should be willing to work together to make the 
utilization and throughput optimal all of the time.
Let us all try to keep things simple and clear.  Too many rules make it 
hard to understand and follow for both new and old  users alike.  Then 
this will add the complexity to the semantics of the rules and various 
interpretations of the rules, and subsequent evolution of 'sneaky' 
methods to get around the rules we have in place...

The 16 minute rule in place allows you to show up as late as minute 16 
after your reservation period starts and you still have a reservation 
slot. if you don't call, post a note on the system, nor indicate you 
will be using the reservation before this time you will loose your 
reservation.
I request users follow this order of hierarchy:
First the user currently enabled on the system can continue his EBL 
session if desired; or the next user scheduled on CORAL, if present and 
ready to write, can come on the system early. Finally if the previous 
mentioned persons can't write then any available RAITH user that enables 
the tool on CORAL has the cancelled/no-show session.
It is only at his or her courtesy if they hand off the system back to 
the previous user that was scheduled if there remains time on the 
reservation slot.  Often Users will exchange time slots at this 
occurrence such as the common one user running late in the cleanroom and 
the following user is ready and present to write.

A little courtesy and cooperation between users will go a long way 
towards increasing throughput on the system.

Secondly if you know a user is coming on time I would expect you to 
cancel the remainder of your run and start unloading at the end of your 
reservation slot,  or beg well in advance for their permission to extend 
into that persons reservation time at their option. I feel it is is also 
important to be flexible with this regard as we all have had both good 
and bad days on the system.  It would be expected that the hand off 
would occur cleanly during the first 15 minute of the reservation.

I feel their is no practical means nor need to add any rules about short 
open times between reservations; but will reconsider if people start 
using this as a ''sneaky'' method to extend their utilization of the 
system in deference to users whom also need access to the system.  
Please be above board and in plain view regarding all reservations, 
cancellations, and hand-offs informally between users should be posted 
to this discussion list.

Your comments are invited.

James Conway


Mark Topinka wrote:

> Ryan has a valid point, too.. sometimes if I have extra things to 
> write I wait to unload my sample until I'm sure the next person is 
> going to show up, and this often means the my reservation bleeds 10 
> minutes over into the next reservation.  But I also don't expect the 
> person before me to be off right at the time I arrive- rather, I 
> expect that they will start unloading when I arrive (at the beginning 
> of my reservation).  I do generally try to start the unload procedure 
> immediately when the next person shows up.  We've never clarified the 
> rules on what a "2pm to 6pm" reservation means- does this mean that 
> the user before you is expected to be off the machine by 2?  and 
> you're expected to be unloaded by 6?  or does it mean that the user 
> before you should unload right at 2 (assuming you show up at 2), and 
> you should unload right at 6 (assuming the next user has shown up)?  I 
> prefer the latter, because then if a user doesn't show up or is late, 
> we don't let the Raith just sit around doing nothing during that 
> time.  The other part of the solution is what James has been 
> advocating for a while now: arranging with the user before you and/or 
> the user after you so that you all load your samples at the same time, 
> and we don't waste 30 minutes on every handoff.    Sorry for the 
> torrent of email here, but I think it's important to have clear rules 
> that everybody understands and is happy with.  -Mark
>
>
> At 01:20 PM 11/16/2004 -0800, Ryan Tu wrote:
>
>> I would like to comment on leaving half an hour between reservations, 
>> which
>> is something I do because in my experience, users are rarely on time for
>> handoff.  I am only allowed 8 hours per 10 day period and to waste 
>> half an
>> hour per session waiting for someone else to unload their sample is a
>> significant waste of reservation time, especially when I have 
>> multiple short
>> 2-hour writes.
>>
>> If we can address the issue of people unloading on time, then I fully 
>> agree
>> with Mark's suggestion.
>>
>> Ryan
>>
>>
>> On 11/16/04 1:01 PM, "Jien Cao" <jiencao@stanford.edu> wrote:
>>
>> > I agree.  Each reservation should probably go immediately
>> > after the previous one, or else leave a 2- or 4-hour gap in
>> > between.
>> >
>> > However, in the case when there are already reservations
>> > ahead and after, how should this work?  The newly reservation
>> > still has to immediately follow the previous one, but is it allowed
>> > to leave a gap between itself and the reservation after (which
>> > has already been there)?  We should probably be flexible with
>> > this case and not force a user to reserve the time he/she doesn't
>> > need...
>> >
>> > Jien
>> > ----- Original Message -----
>> > From: "Scott D. Andrews" <sandrew@stanford.edu>
>> > To: <raith@snf.stanford.edu>
>> > Sent: Tuesday, November 16, 2004 12:43 PM
>> > Subject: Re: Responses and suggestions on Raith reservation policy
>> >
>> >
>> >>
>> >>> penalizes people who can legitimately get a write done quickly!?   I
>> >>> believe my proposal ("you are not allowed to leave any 1/2 or 1 hour
>> >>> block
>> >>> between your reservation and another user's reservation.  And no
>> >>> reservation shorter than 2 hours.") solves the sneaky-half 
>> reservations
>> >>> problem that some people are doing,
>> >>
>> >> I second Mark's suggestion.  I think it is a very good idea.
>> >>
>> >> -Scott
>> >>
>> >>
>> >>
>> >>
>> >
>> >
>


Greetings:

Interesting discussion on this topic thread. 
I feel the fewer rules the better, and resist all suggestions to add more rules as people should be willing to work together to make the utilization and throughput optimal all of the time.
Let us all try to keep things simple and clear.  Too many rules make it hard to understand and follow for both new and old  users alike.  Then this will add the complexity to the semantics of the rules and various interpretations of the rules, and subsequent evolution of 'sneaky' methods to get around the rules we have in place...

The 16 minute rule in place allows you to show up as late as minute 16 after your reservation period starts and you still have a reservation slot. if you don't call, post a note on the system, nor indicate you will be using the reservation before this time you will loose your reservation.
I request users follow this order of hierarchy:
First the user currently enabled on the system can continue his EBL session if desired; or the next user scheduled on CORAL, if present and ready to write, can come on the system early. Finally if the previous mentioned persons can't write then any available RAITH user that enables the tool on CORAL has the cancelled/no-show session.
It is only at his or her courtesy if they hand off the system back to the previous user that was scheduled if there remains time on the reservation slot.  Often Users will exchange time slots at this occurrence such as the common one user running late in the cleanroom and the following user is ready and present to write.

A little courtesy and cooperation between users will go a long way towards increasing throughput on the system.

Secondly if you know a user is coming on time I would expect you to cancel the remainder of your run and start unloading at the end of your reservation slot,  or beg well in advance for their permission to extend into that persons reservation time at their option. I feel it is is also important to be flexible with this regard as we all have had both good and bad days on the system.  It would be expected that the hand off would occur cleanly during the first 15 minute of the reservation.

I feel their is no practical means nor need to add any rules about short open times between reservations; but will reconsider if people start using this as a ''sneaky'' method to extend their utilization of the system in deference to users whom also need access to the system.  Please be above board and in plain view regarding all reservations, cancellations, and hand-offs informally between users should be posted to this discussion list.

Your comments are invited.

James Conway


Mark Topinka wrote:
Ryan has a valid point, too.. sometimes if I have extra things to write I wait to unload my sample until I'm sure the next person is going to show up, and this often means the my reservation bleeds 10 minutes over into the next reservation.  But I also don't expect the person before me to be off right at the time I arrive- rather, I expect that they will start unloading when I arrive (at the beginning of my reservation).  I do generally try to start the unload procedure immediately when the next person shows up.  We've never clarified the rules on what a "2pm to 6pm" reservation means- does this mean that the user before you is expected to be off the machine by 2?  and you're expected to be unloaded by 6?  or does it mean that the user before you should unload right at 2 (assuming you show up at 2), and you should unload right at 6 (assuming the next user has shown up)?  I prefer the latter, because then if a user doesn't show up or is late, we don't let the Raith just sit around doing nothing during that time.  The other part of the solution is what James has been advocating for a while now: arranging with the user before you and/or the user after you so that you all load your samples at the same time, and we don't waste 30 minutes on every handoff.    Sorry for the torrent of email here, but I think it's important to have clear rules that everybody understands and is happy with.  -Mark


At 01:20 PM 11/16/2004 -0800, Ryan Tu wrote:
I would like to comment on leaving half an hour between reservations, which
is something I do because in my experience, users are rarely on time for
handoff.  I am only allowed 8 hours per 10 day period and to waste half an
hour per session waiting for someone else to unload their sample is a
significant waste of reservation time, especially when I have multiple short
2-hour writes.

If we can address the issue of people unloading on time, then I fully agree
with Mark's suggestion.

Ryan


On 11/16/04 1:01 PM, "Jien Cao" <jiencao@stanford.edu> wrote:

> I agree.  Each reservation should probably go immediately
> after the previous one, or else leave a 2- or 4-hour gap in
> between.
>
> However, in the case when there are already reservations
> ahead and after, how should this work?  The newly reservation
> still has to immediately follow the previous one, but is it allowed
> to leave a gap between itself and the reservation after (which
> has already been there)?  We should probably be flexible with
> this case and not force a user to reserve the time he/she doesn't
> need...
>
> Jien
> ----- Original Message -----
> From: "Scott D. Andrews" <sandrew@stanford.edu>
> To: <raith@snf.stanford.edu>
> Sent: Tuesday, November 16, 2004 12:43 PM
> Subject: Re: Responses and suggestions on Raith reservation policy
>
>
>>
>>> penalizes people who can legitimately get a write done quickly!?   I
>>> believe my proposal ("you are not allowed to leave any 1/2 or 1 hour
>>> block
>>> between your reservation and another user's reservation.  And no
>>> reservation shorter than 2 hours.") solves the sneaky-half reservations
>>> problem that some people are doing,
>>
>> I second Mark's suggestion.  I think it is a very good idea.
>>
>> -Scott
>>
>>
>>
>>
>
>