There are logic bugs where you get a syntax error.
Design flaws where is does not do as you expected.
Then the "Features" that don't please the user and don't function as they expected.
You raise three points which I'd like to answer:-
1. Syntax errors?? How do you compile code with syntax errors? Surely you are talking about logic bugs where people like us find them and help the devs fix them. Seems pretty normal and expected to me.
2. Design Flaws?? In whose opinion - yours?? You have a perfect right along with everyone else to make suggestions for design changes in either Problems and Bug Reports or the Wishlist. I've observed the process work quite well where the person reporting the problem shows that he has done a bit of basic research into the causes of the problem and can clearly identify the causes of the problem. Just stating "It's a bug" or "It's bad design" doesn't actually make it so. Also you should figure out if the design flaw is BOINC only or EAH only or in fact a combination of the two and report in the correct forum.
3. Bad or Non-functional features?? Once again, a matter of opinion and once again, research it and fully report it. As far as BOINC features are concerned, why don't you download a copy of the code and develop a patch to fix the feature or improve its functionality? That'd be far more helpful than just being a winger. The devs are very receptive to well written enhancements. Just look at what JM7 has done by deciding to get his hands dirty in the code.
Actually I still think you owe JM7 an apology.
Quote:
It seems to be doing well after I cleared the LTD and now has a a few WU's from each project.
Translation - it seems to be doing well now that I've told it to ignore my own choices for resource shares. It should have been smart enough in the first place to have figured out that it needs to do "What I think I want" rather than "what I actually said I want". Isn't it a piece of stupid, bug ridden software.
Quote:
Now if the LTD is the hidden driving force then the user should get this explained in the help or on the BOINC site. The "Requesting 0 seconds" message is no reason to make the user think there is a problem.
You say you write code for a living. Code writers write code - Documenters write documentation. It would be quite unusual to get daily updates in the documentation as the daily code changes roll out. Documentation invariably lags well behind and because of the many changes in the area of STD/LTD that have happened in the last couple of months it would be an absolute miracle (and a complete waste of time) if the documentation were right up to the minute. Tomorrow the code will be different.
However the BOINC project is indeed fortunate to have a dedicated team of documenters headed by Paul Buck who are doing a marvellous job with the BOINC Wiki. Once again, your "broad brush" and unjustified criticism has failed to notice how much of the intricacies of BOINC are actually documented in the Wiki. Go and read it some time and find out how well the concept of "debt" is explained there.
Quote:
Now the average user should have about 4 things to set up. 1) network connection 2) Select projects to do 3) set network connection time / cache size 4) fiddle with the percentages to give each project.
I am sure the average user could not follow the explanation of what is really happening nor why it is not a bug.
Statements like this suggest that you really haven't given the system a fair appraisal and you are certainly underestimating the capabilities of the average user. The 4 steps you suggest are really all that the average user needs to do. What more blindingly difficult steps are there? In fact the user can choose the defaults for most preferences and leave out step 3 and if the user has previously browsed the net, there shouldn't be much if anything to do in step 1.
The average user doesn't need to be troubled by STD/LTD and therefore can be completely oblivious to the underlying mechanisms. All the average user needs to do is set a sane value for "connect to network", stick to what is decided regarding resource shares and then have the good sense to leave BOINC alone to follow the orders. If you must fiddle, then it's your duty to understand the consequences of your own insanity. You can hardly blame BOINC for that.
As of (4.72? 5.1?) a project that is not supplying work will not have its LTDebt increased. However, if the client is not asking for work because of some sort of overload, then the LTDebt will be increased after the current deferral is finished.
John,
That is an excellent solution and should fix the problems of projects with extended outages building up large positive debt when they can't supply anyway. Sounds like a very sane way to go.
Thanks for all your hard work. It is appreciated!!
Now if the LTD is the hidden driving force then the user should get this explained in the help or on the BOINC site. The "Requesting 0 seconds" message is no reason to make the user think there is a problem.
You say you write code for a living. Code writers write code - Documenters write documentation. It would be quite unusual to get daily updates in the documentation as the daily code changes roll out. Documentation invariably lags well behind and because of the many changes in the area of STD/LTD that have happened in the last couple of months it would be an absolute miracle (and a complete waste of time) if the documentation were right up to the minute. Tomorrow the code will be different.
Or the message. I am trying to figure out how to get the new message sequence into the Wiki. ;)
It's no longer Requesting 0 seconds of work, returning 0 results.
It's separated into parts:
2005-09-13 01:42:04 [Einstein@Home] Requesting 43200 seconds of new work
2005-09-13 01:42:59 [BOINC alpha test] Reporting 1 results
2005-09-13 01:49:24 [SETI@home] Note: not requesting new work or reporting results
As you can see, the messaging has been changed completely (this is in 5.1.x Alpha, so it's easily expected to be in 5.2.0 release client). I'm just wondering if it needs a completely new message in the Wiki, or that we can place it under the old ones and slowly change them over to the new ones. ;)
I've been reading the Wiki a bit of late and all I can say is that you guys deserve a medal for getting as far as you have got. It always pisses me off when people complain about lack of documentation for something changing as fast as BOINC is. There's no lack of documentation but you do have to make allowances for the fact that there are many rapid changes of the sort you mention that make the job exceedingly difficult to cover it all and get it all correct.
Anyhow please be aware that there are many of us who understand the difficulties you are working under and who really appreciate the quality of the product you are putting out there for our benefit. My sincere thanks to you, Paul and the rest of the team.
Thank you Gary. Yet seeing how eloquent you write above, how you can explain certain parts to users here and in other threads, I've taken the liberty to point Paul in your direction. We can always use more help in explaining what an error message means.
Most people run screaming into the night when the word documentation is mentioned ... :)
And Gary, thank you for the kind words. It even gets worse when you consider that we have to figure out many things from the code and messages. Though I do have a couple people from projects that have asked for accounts, well, they have a day job too ...
The only redeaming feature is that though many parts do change, there is enough stability that we can, mostly, look to the edges where there is nothing at all.
If anyone cares, the old site is still there and you can look and see that most all of it came over nicely and is still available (the web site owner's manual is still patchy ... )
Anyway, Gary, if you do have suggestions on what you read, we do take guest input ... and some have been saddled with an account without much notice ...
Thanks for the "Vote of Confidence" in my ability to explain some of the things I do understand. I do enjoy challenges which is exactly what I see when people ask questions about something I understand well. However there are probably just as many things that I don't understand well enough and so I can easily avoid them by passing that thread by without answering in those cases :).
Its probably best if I send a lengthier response in private email.
no further work since upgrade to boinc 4.45. I coulnd't find a useful tip in this thread. For climateproject for example I have allocated 4.3 GB disk space but still not sufficient?
29/09/2005 06:50:55|SETI@home|Deferring communication with project for 16 hours, 59 minutes, and 57 seconds
29/09/2005 06:52:31|climateprediction.net|Deferring communication with project for 13 hours, 57 minutes, and 11 seconds
29/09/2005 06:52:31|LHC@home|Deferring communication with project for 13 hours, 58 minutes, and 23 seconds
29/09/2005 06:52:36||Insufficient work; requesting more
29/09/2005 07:33:15|Einstein@Home|Deferring communication with project for 13 hours, 59 minutes, and 57 seconds
29/09/2005 07:48:15||request_reschedule_cpus: project op
29/09/2005 07:48:15|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
29/09/2005 07:48:15|climateprediction.net|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:16|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
29/09/2005 07:48:16|climateprediction.net|Message from server: No work sent
29/09/2005 07:48:16|climateprediction.net|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:16|climateprediction.net|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:16|climateprediction.net|No work from project
29/09/2005 07:48:17|climateprediction.net|Deferring communication with project for 23 hours, 59 minutes, and 59 seconds
29/09/2005 07:48:19||request_reschedule_cpus: project op
29/09/2005 07:48:21|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
29/09/2005 07:48:21|Einstein@Home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:22||request_reschedule_cpus: project op
29/09/2005 07:48:22|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
29/09/2005 07:48:23|Einstein@Home|Message from server: No work sent
29/09/2005 07:48:23|Einstein@Home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:23|Einstein@Home|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:23|Einstein@Home|No work from project
29/09/2005 07:48:24|LHC@home|Sending scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi
29/09/2005 07:48:24|LHC@home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:25|LHC@home|Scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi succeeded
29/09/2005 07:48:25|LHC@home|Message from server: No work sent
29/09/2005 07:48:25|LHC@home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:25|LHC@home|Message from server: No disk space (you must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:25|LHC@home|No work from project
29/09/2005 07:48:26|Einstein@Home|Deferring communication with project for 23 hours, 59 minutes, and 56 seconds
29/09/2005 07:48:26|LHC@home|Deferring communication with project for 23 hours, 59 minutes, and 58 seconds
29/09/2005 07:48:26||request_reschedule_cpus: project op
29/09/2005 07:48:28||request_reschedule_cpus: project op
29/09/2005 07:48:29|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
29/09/2005 07:48:29|SETI@home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:31|SETI@home|Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
29/09/2005 07:48:31|SETI@home|Message from server: No work sent
29/09/2005 07:48:31|SETI@home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:31|SETI@home|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:31|SETI@home|No work from project
29/09/2005 07:48:32|SETI@home|Deferring communication with project for 23 hours, 59 minutes, and 58 seconds
29/09/2005 07:52:36||Insufficient work; requesting more
no further work since upgrade to boinc 4.45. I coulnd't find a useful tip in this thread. For climateproject for example I have allocated 4.3 GB disk space but still not sufficient?
Instead of looking in this thread why not search for the key parts of the error message in the Wiki. Just type them into the search box on the main page and read all about it. In this case however it may be even simpler. You say you have plenty free space but unfortunately BOINC doesn't see it. Is the free space on the same partition, is there some sort of permissions problem with BOINC using the free space, or is there a stuff-up with the way you have set up your memory and disk usage general prefs on the website? These three things are probably the first to check. There are five preference settings involved. Perhaps you would like to tell us the values you have set for those?
I don't actually use 4.45 so if these problems are entirely due to the upgrade then maybe there is some particular bug in 4.45 that I'm not aware of. I've seen good reports about 4.72 so maybe you could try uninstalling 4.45 and then installing 4.72 and see if the problem goes away. Please realise that 4.72 is a development version and not guaranteed to be bug free, use at your own risk, etc, but you can really say exactly the same about 4.45 :). Someone is bound to chime in about 5.1.4 so maybe you could try that after 4.72 if you still get no joy. Same caveats apply!!
Someone is bound to chime in about 5.1.4 so maybe you could try that after 4.72 if you still get no joy.
5.1.5 at the time of writing, but I won't chime it in. It will most probably give the same error message.
It is something to do with the disk settings at the general preferences, plus your physical disk drive's free space. So for me, please state how big the drive is, if it is a partition, if it really has enough space free etc.
@Gary, good to see you didn't run off screaming into the night. Even if you'd only be correcting the English or legibility on some of the pages in the Wiki, it would help out. Hopefully Paul will forward the lengthy email to me, then I'll return to you what I wrote to him yesterday. :)
RE: There are logic bugs
)
You raise three points which I'd like to answer:-
1. Syntax errors?? How do you compile code with syntax errors? Surely you are talking about logic bugs where people like us find them and help the devs fix them. Seems pretty normal and expected to me.
2. Design Flaws?? In whose opinion - yours?? You have a perfect right along with everyone else to make suggestions for design changes in either Problems and Bug Reports or the Wishlist. I've observed the process work quite well where the person reporting the problem shows that he has done a bit of basic research into the causes of the problem and can clearly identify the causes of the problem. Just stating "It's a bug" or "It's bad design" doesn't actually make it so. Also you should figure out if the design flaw is BOINC only or EAH only or in fact a combination of the two and report in the correct forum.
3. Bad or Non-functional features?? Once again, a matter of opinion and once again, research it and fully report it. As far as BOINC features are concerned, why don't you download a copy of the code and develop a patch to fix the feature or improve its functionality? That'd be far more helpful than just being a winger. The devs are very receptive to well written enhancements. Just look at what JM7 has done by deciding to get his hands dirty in the code.
Actually I still think you owe JM7 an apology.
Translation - it seems to be doing well now that I've told it to ignore my own choices for resource shares. It should have been smart enough in the first place to have figured out that it needs to do "What I think I want" rather than "what I actually said I want". Isn't it a piece of stupid, bug ridden software.
You say you write code for a living. Code writers write code - Documenters write documentation. It would be quite unusual to get daily updates in the documentation as the daily code changes roll out. Documentation invariably lags well behind and because of the many changes in the area of STD/LTD that have happened in the last couple of months it would be an absolute miracle (and a complete waste of time) if the documentation were right up to the minute. Tomorrow the code will be different.
However the BOINC project is indeed fortunate to have a dedicated team of documenters headed by Paul Buck who are doing a marvellous job with the BOINC Wiki. Once again, your "broad brush" and unjustified criticism has failed to notice how much of the intricacies of BOINC are actually documented in the Wiki. Go and read it some time and find out how well the concept of "debt" is explained there.
Statements like this suggest that you really haven't given the system a fair appraisal and you are certainly underestimating the capabilities of the average user. The 4 steps you suggest are really all that the average user needs to do. What more blindingly difficult steps are there? In fact the user can choose the defaults for most preferences and leave out step 3 and if the user has previously browsed the net, there shouldn't be much if anything to do in step 1.
The average user doesn't need to be troubled by STD/LTD and therefore can be completely oblivious to the underlying mechanisms. All the average user needs to do is set a sane value for "connect to network", stick to what is decided regarding resource shares and then have the good sense to leave BOINC alone to follow the orders. If you must fiddle, then it's your duty to understand the consequences of your own insanity. You can hardly blame BOINC for that.
Cheers,
Gary.
RE: As of (4.72? 5.1?) a
)
John,
That is an excellent solution and should fix the problems of projects with extended outages building up large positive debt when they can't supply anyway. Sounds like a very sane way to go.
Thanks for all your hard work. It is appreciated!!
Cheers,
Gary.
RE: RE: Now if the LTD is
)
Or the message. I am trying to figure out how to get the new message sequence into the Wiki. ;)
It's no longer Requesting 0 seconds of work, returning 0 results.
It's separated into parts:
2005-09-13 01:42:04 [Einstein@Home] Requesting 43200 seconds of new work
2005-09-13 01:42:59 [BOINC alpha test] Reporting 1 results
2005-09-13 01:49:24 [SETI@home] Note: not requesting new work or reporting results
As you can see, the messaging has been changed completely (this is in 5.1.x Alpha, so it's easily expected to be in 5.2.0 release client). I'm just wondering if it needs a completely new message in the Wiki, or that we can place it under the old ones and slowly change them over to the new ones. ;)
For links to the Wiki, there can be only one. :)
(Jim's next comment will no doubt be that he can't edit the Wiki. It's on invitation only.)
@Ageless I've been reading
)
@Ageless
I've been reading the Wiki a bit of late and all I can say is that you guys deserve a medal for getting as far as you have got. It always pisses me off when people complain about lack of documentation for something changing as fast as BOINC is. There's no lack of documentation but you do have to make allowances for the fact that there are many rapid changes of the sort you mention that make the job exceedingly difficult to cover it all and get it all correct.
Anyhow please be aware that there are many of us who understand the difficulties you are working under and who really appreciate the quality of the product you are putting out there for our benefit. My sincere thanks to you, Paul and the rest of the team.
Cheers,
Gary.
Thank you Gary. Yet seeing
)
Thank you Gary. Yet seeing how eloquent you write above, how you can explain certain parts to users here and in other threads, I've taken the liberty to point Paul in your direction. We can always use more help in explaining what an error message means.
But how do we know he is even
)
But how do we know he is even interested?
Most people run screaming into the night when the word documentation is mentioned ... :)
And Gary, thank you for the kind words. It even gets worse when you consider that we have to figure out many things from the code and messages. Though I do have a couple people from projects that have asked for accounts, well, they have a day job too ...
The only redeaming feature is that though many parts do change, there is enough stability that we can, mostly, look to the edges where there is nothing at all.
If anyone cares, the old site is still there and you can look and see that most all of it came over nicely and is still available (the web site owner's manual is still patchy ... )
Anyway, Gary, if you do have suggestions on what you read, we do take guest input ... and some have been saddled with an account without much notice ...
Jord and Paul, Thanks for
)
Jord and Paul,
Thanks for the "Vote of Confidence" in my ability to explain some of the things I do understand. I do enjoy challenges which is exactly what I see when people ask questions about something I understand well. However there are probably just as many things that I don't understand well enough and so I can easily avoid them by passing that thread by without answering in those cases :).
Its probably best if I send a lengthier response in private email.
Cheers,
Gary.
no further work since upgrade
)
no further work since upgrade to boinc 4.45. I coulnd't find a useful tip in this thread. For climateproject for example I have allocated 4.3 GB disk space but still not sufficient?
29/09/2005 06:50:55|SETI@home|Deferring communication with project for 16 hours, 59 minutes, and 57 seconds
29/09/2005 06:52:31|climateprediction.net|Deferring communication with project for 13 hours, 57 minutes, and 11 seconds
29/09/2005 06:52:31|LHC@home|Deferring communication with project for 13 hours, 58 minutes, and 23 seconds
29/09/2005 06:52:36||Insufficient work; requesting more
29/09/2005 07:33:15|Einstein@Home|Deferring communication with project for 13 hours, 59 minutes, and 57 seconds
29/09/2005 07:48:15||request_reschedule_cpus: project op
29/09/2005 07:48:15|climateprediction.net|Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
29/09/2005 07:48:15|climateprediction.net|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:16|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
29/09/2005 07:48:16|climateprediction.net|Message from server: No work sent
29/09/2005 07:48:16|climateprediction.net|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:16|climateprediction.net|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:16|climateprediction.net|No work from project
29/09/2005 07:48:17|climateprediction.net|Deferring communication with project for 23 hours, 59 minutes, and 59 seconds
29/09/2005 07:48:19||request_reschedule_cpus: project op
29/09/2005 07:48:21|Einstein@Home|Sending scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
29/09/2005 07:48:21|Einstein@Home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:22||request_reschedule_cpus: project op
29/09/2005 07:48:22|Einstein@Home|Scheduler request to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
29/09/2005 07:48:23|Einstein@Home|Message from server: No work sent
29/09/2005 07:48:23|Einstein@Home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:23|Einstein@Home|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:23|Einstein@Home|No work from project
29/09/2005 07:48:24|LHC@home|Sending scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi
29/09/2005 07:48:24|LHC@home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:25|LHC@home|Scheduler request to http://lhcathome-sched1.cern.ch/scheduler/cgi succeeded
29/09/2005 07:48:25|LHC@home|Message from server: No work sent
29/09/2005 07:48:25|LHC@home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:25|LHC@home|Message from server: No disk space (you must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:25|LHC@home|No work from project
29/09/2005 07:48:26|Einstein@Home|Deferring communication with project for 23 hours, 59 minutes, and 56 seconds
29/09/2005 07:48:26|LHC@home|Deferring communication with project for 23 hours, 59 minutes, and 58 seconds
29/09/2005 07:48:26||request_reschedule_cpus: project op
29/09/2005 07:48:28||request_reschedule_cpus: project op
29/09/2005 07:48:29|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
29/09/2005 07:48:29|SETI@home|Requesting 8640 seconds of work, returning 0 results
29/09/2005 07:48:31|SETI@home|Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
29/09/2005 07:48:31|SETI@home|Message from server: No work sent
29/09/2005 07:48:31|SETI@home|Message from server: (there was work but you don't have enough disk space allocated)
29/09/2005 07:48:31|SETI@home|Message from server: No disk space (YOU must free 1731.7 MB before BOINC gets space). Review preferences for minimum disk free space allowed.
29/09/2005 07:48:31|SETI@home|No work from project
29/09/2005 07:48:32|SETI@home|Deferring communication with project for 23 hours, 59 minutes, and 58 seconds
29/09/2005 07:52:36||Insufficient work; requesting more
RE: no further work since
)
Instead of looking in this thread why not search for the key parts of the error message in the Wiki. Just type them into the search box on the main page and read all about it. In this case however it may be even simpler. You say you have plenty free space but unfortunately BOINC doesn't see it. Is the free space on the same partition, is there some sort of permissions problem with BOINC using the free space, or is there a stuff-up with the way you have set up your memory and disk usage general prefs on the website? These three things are probably the first to check. There are five preference settings involved. Perhaps you would like to tell us the values you have set for those?
I don't actually use 4.45 so if these problems are entirely due to the upgrade then maybe there is some particular bug in 4.45 that I'm not aware of. I've seen good reports about 4.72 so maybe you could try uninstalling 4.45 and then installing 4.72 and see if the problem goes away. Please realise that 4.72 is a development version and not guaranteed to be bug free, use at your own risk, etc, but you can really say exactly the same about 4.45 :). Someone is bound to chime in about 5.1.4 so maybe you could try that after 4.72 if you still get no joy. Same caveats apply!!
Good luck!!
Cheers,
Gary.
RE: Someone is bound to
)
5.1.5 at the time of writing, but I won't chime it in. It will most probably give the same error message.
It is something to do with the disk settings at the general preferences, plus your physical disk drive's free space. So for me, please state how big the drive is, if it is a partition, if it really has enough space free etc.
@Gary, good to see you didn't run off screaming into the night. Even if you'd only be correcting the English or legibility on some of the pages in the Wiki, it would help out. Hopefully Paul will forward the lengthy email to me, then I'll return to you what I wrote to him yesterday. :)