More merger proposals

Message boards : Number crunching : More merger proposals
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
Profile skgiven
       
Avatar

Send message
Joined: 7 Sep 10
Posts: 453
Credit: 945,109
RAC: 0
Message 5318 - Posted: 2 Jul 2017, 13:54:50 UTC - in response to Message 5316.  
Last modified: 2 Jul 2017, 14:15:43 UTC

You can use one app to run numerous different types of research. If you're a researcher it might make sense to simply rename the app for each different piece of scientific research you perform & deprecate the old app.
If you need to tweak the app a bit, define parameters/limitations, or group research by researcher, it makes some sense. This might be the case for some active projects.
Other projects use one or two apps and define parameters within the tasks. They indicate the research using a WU/Task naming system, to allow internal badge/research groupings.
These are just two ways of doing things. There is no fixed way and management of 'not my project' isn't my call, it's their choice.
If projects want to launch/re-launch apps to promote themselves that's their call too. Some projects have been doing just that for many years (and long before wuprop had a badge system).
How Sebastien chooses to handle this is down to him (merge/don't merge, list/don't list...) and while he might take on board our views there might be project management constraints at play here too.
Our choice is to crunch or not to crunch whatever is available and in whatever form it turns up.

---

The only reason I attached to some of these projects was to reach 100h or some higher target. In hind sight that was a waste of my time.
Another negative for me is that I'm no longer attached to some of the projects I reached my goals for, and will have to reattach just to reach new targets (sometimes for just a days worth of crunching).
So long as the apps I want to upgrade are listed correctly (against the presently available project) that should at least be doable.

Looking at my projects/app list I think this is a work in progress. Duplicates:

LHC@home LHCb Simulation 296.78 0.00
LHC@home CMS Simulation 264.63 0.00
LHC@home Theory Simulation 258.33 0.00

lhcathome-dev Theory Simulation 569.85 0.00
lhcathome-dev CMS Simulation 289.78 0.00
lhcathome-dev LHCb Simulation 168.80 0.00


What would be helpful is a list of what apps were merged and for what projects and a list of what's left to merge, before some of us start changing what we crunch again.
ID: 5318 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin
       

Send message
Joined: 22 Aug 16
Posts: 447
Credit: 2,089,953
RAC: 709
Message 5319 - Posted: 2 Jul 2017, 19:45:52 UTC - in response to Message 5318.  

You can use one app to run numerous different types of research. If you're a researcher it might make sense to simply rename the app for each different piece of scientific research you perform & deprecate the old app.
If you need to tweak the app a bit, define parameters/limitations, or group research by researcher, it makes some sense. This might be the case for some active projects.
Other projects use one or two apps and define parameters within the tasks. They indicate the research using a WU/Task naming system, to allow internal badge/research groupings.
These are just two ways of doing things. There is no fixed way and management of 'not my project' isn't my call, it's their choice.
If projects want to launch/re-launch apps to promote themselves that's their call too. Some projects have been doing just that for many years (and long before wuprop had a badge system).
How Sebastien chooses to handle this is down to him (merge/don't merge, list/don't list...) and while he might take on board our views there might be project management constraints at play here too.
Our choice is to crunch or not to crunch whatever is available and in whatever form it turns up.

---

The only reason I attached to some of these projects was to reach 100h or some higher target. In hind sight that was a waste of my time.
Another negative for me is that I'm no longer attached to some of the projects I reached my goals for, and will have to reattach just to reach new targets (sometimes for just a days worth of crunching).
So long as the apps I want to upgrade are listed correctly (against the presently available project) that should at least be doable.

Looking at my projects/app list I think this is a work in progress. Duplicates:

LHC@home LHCb Simulation 296.78 0.00
LHC@home CMS Simulation 264.63 0.00
LHC@home Theory Simulation 258.33 0.00

lhcathome-dev Theory Simulation 569.85 0.00
lhcathome-dev CMS Simulation 289.78 0.00
lhcathome-dev LHCb Simulation 168.80 0.00


What would be helpful is a list of what apps were merged and for what projects and a list of what's left to merge, before some of us start changing what we crunch again.


LHC and LHC dev are completely different websites, different sets of project preferences. Where does the line get drawn? Might as well group all apps from a project into one. Just leave it as the project admins set it.

Of the WUProp users I've looked at no one has a star (100hr mine)for every app they have listed. There will always be apps we don't get a star for or as high of a star before it ends. If the admin names them all different then I'd say leave it. We're all here to get stars anyway.
ID: 5319 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Skivelitis2
     
Avatar

Send message
Joined: 8 Nov 14
Posts: 26
Credit: 2,821,375
RAC: 69
Message 5320 - Posted: 2 Jul 2017, 21:22:04 UTC
Last modified: 2 Jul 2017, 21:26:45 UTC

Be careful when you say all or KPX will think you are speaking for him! LOL! Yes, we all (well most of us) love to get the stars. Myself, I generally only crunch projects that grant badges unless my team is involved in a challenge. If I have a project goal that will also help towards an improved star here so much the better. The original intention of this project was to encourage crunching across the board regardless of credit granted and has done an amazing job achieving this goal. The statistics shown under the results tab are an immense help to me when deciding whether my very limited hardware is up to the task for certain apps. For this reason, I wish every BOINC user would run this project and contribute to what would be a greatly enhanced database for all. Damn, there's that word again.

BTW: Couldn't agree more with skgiven whereas it would have been nice had the merger occurred much sooner. In sebastien's defense however, nobody knew how everything would shake out at GCC. Better to get it right later than never.
ID: 5320 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Steve Hawker*
       

Send message
Joined: 31 Oct 12
Posts: 125
Credit: 655,634
RAC: 0
Message 5321 - Posted: 3 Jul 2017, 1:33:41 UTC - in response to Message 5320.  

nobody knew how everything would shake out at GCC.

Not sure what you mean here. Goofy made it abundantly clear that he was moving everything to GCC and indeed he did. Pretty sure that meets the definition of "knowing how things will shake out"

I'm fine with any merger so long as the "system" is consistent. Goofyxgrid moved to GCC so it would be consistent to lump hours together.

The one that confuses me is WCC Beta (see thread passim)
ID: 5321 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin
       

Send message
Joined: 22 Aug 16
Posts: 447
Credit: 2,089,953
RAC: 709
Message 5322 - Posted: 3 Jul 2017, 2:00:57 UTC - in response to Message 5320.  

Be careful when you say all or KPX will think you are speaking for him! LOL! Yes, we all (well most of us) love to get the stars. Myself, I generally only crunch projects that grant badges unless my team is involved in a challenge. If I have a project goal that will also help towards an improved star here so much the better. The original intention of this project was to encourage crunching across the board regardless of credit granted and has done an amazing job achieving this goal. The statistics shown under the results tab are an immense help to me when deciding whether my very limited hardware is up to the task for certain apps. For this reason, I wish every BOINC user would run this project and contribute to what would be a greatly enhanced database for all. Damn, there's that word again.

BTW: Couldn't agree more with skgiven whereas it would have been nice had the merger occurred much sooner. In sebastien's defense however, nobody knew how everything would shake out at GCC. Better to get it right later than never.


I agree. There are projects I've run and told my team about just because I hear about it here and gather WUProp hours. Recent ones would be ODLK and Xansons.
ID: 5322 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile skgiven
       
Avatar

Send message
Joined: 7 Sep 10
Posts: 453
Credit: 945,109
RAC: 0
Message 5323 - Posted: 3 Jul 2017, 8:22:45 UTC - in response to Message 5321.  
Last modified: 3 Jul 2017, 8:25:36 UTC

The one that confuses me is WCC Beta (see thread passim)

Ah, there's that elephant. Was wandering which room I left it in.

When WCG are Beta testing a new app, they conceal the new projects identity, but we have 'Beta' and 'Beta Test' listed.
When they are testing an existing app/new version they call it something like 'Beta FightAIDS@Home'.
IMO however, all the WCG Beta's should be pooled into the one WCG Beta app because at WCG they treat Beta testing as a sub-project and have a badge for Beta contributions.
What we have is a representation of the different apps (where known) but it's a bit of a mess:
    World Community Grid Beta Test 658.73
    World Community Grid Beta - FightAIDS@Home 224.78
    World Community Grid Beta - FightAIDS@Home - VINA 108.95

    World Community Grid Beta 73.05
    World Community Grid Beta - FightAIDS@Home - AutoDock 37.08
    World Community Grid Beta - The Clean Energy Project - Phase 2 31.22
    World Community Grid Beta - Say No to Schistosoma 1.63 0.00


Most of us would try to reach at least 100h for any new app we chose to crunch, so it's not by choice that we end up with lots of apps with less than 100h that will never go anywhere. The tasks were simply not available, even for big projects.
All the above Beta's have been inactive for a long time and all but one of the 45 apps I have with <100h are now deprecated as are some of the research lines/projects/sub-projects.

If all the WCG Beta apps were merged I would lose two lower apps but surpass 1000h for the one WCG Beta star and be able to add to it in the future. Stats for flyby night Betas are meaningless anyway. By the time you get the stats the apps are gone.

A bit O/T but I don't need to see 44 deprecated apps hanging off the bottom of my list every time I log in.
A Hide deprecated apps with <100h option would be nice.

ID: 5323 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin
       

Send message
Joined: 22 Aug 16
Posts: 447
Credit: 2,089,953
RAC: 709
Message 5324 - Posted: 3 Jul 2017, 11:58:58 UTC - in response to Message 5323.  

Most of us would try to reach at least 100h for any new app we chose to crunch, so it's not by choice that we end up with lots of apps with less than 100h that will never go anywhere. The tasks were simply not available, even for big projects.
All the above Beta's have been inactive for a long time and all but one of the 45 apps I have with <100h are now deprecated as are some of the research lines/projects/sub-projects.

A bit O/T but I don't need to see 44 deprecated apps hanging off the bottom of my list every time I log in.
A Hide deprecated apps with <100h option would be nice.


My point mentioning the apps less than 100hr is that it's just a part of WUProp. We all have those apps. It happens. Sometimes we don't get to it in time, others we just can't get enough work. We shouldn't be trying to merge them to get a higher star or a star at all. Again, where does the merging end if we go down this path.
ID: 5324 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile skgiven
       
Avatar

Send message
Joined: 7 Sep 10
Posts: 453
Credit: 945,109
RAC: 0
Message 5325 - Posted: 3 Jul 2017, 17:01:34 UTC - in response to Message 5324.  

Most of us would try to reach at least 100h for any new app we chose to crunch, so it's not by choice that we end up with lots of apps with less than 100h that will never go anywhere. The tasks were simply not available, even for big projects.
All the above Beta's have been inactive for a long time and all but one of the 45 apps I have with <100h are now deprecated as are some of the research lines/projects/sub-projects.

A bit O/T but I don't need to see 44 deprecated apps hanging off the bottom of my list every time I log in.
A Hide deprecated apps with <100h option would be nice.


My point mentioning the apps less than 100hr is that it's just a part of WUProp. We all have those apps. It happens. Sometimes we don't get to it in time, others we just can't get enough work. We shouldn't be trying to merge them to get a higher star or a star at all. Again, where does the merging end if we go down this path.

The purpose of WUProp is to produce meaningful stats & the reward is the fun getting stars/milestones.
There is no purpose to a Beta app that turns up for a few hours & a few users get the odd work unit or two.
Unless an app is around long enough for the stats to mean anything, be acted upon, it's junk to WUProp & WUProp crunchers.
Why keep a list of apps with <100h when the app has been deprecated & often the project/science finished.
I'm saying merge what can be merged or delete that which has no meaning, function or value.
If there is a good reason to keep a list of apps that are deprecated with <100h run time, then OK, but I don't see any point in it.

- Just my view on the matter & it's no biggie whatever happens, or doesn't.
ID: 5325 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
zombie67 [MM]
             
Avatar

Send message
Joined: 30 Mar 10
Posts: 227
Credit: 9,069,863
RAC: 1,039
Message 5353 - Posted: 8 Jul 2017, 15:12:53 UTC

FWIW, lhcathome-dev is still an active project, issuing work. IMO, that means the apps should stay separate too.
Reno, NV
Team: SETI.USA
ID: 5353 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile [AF>WildWildWest] Sébastie...
     
Project administrator
Avatar

Send message
Joined: 28 Mar 10
Posts: 2870
Credit: 538,395
RAC: 132
Message 5359 - Posted: 9 Jul 2017, 9:22:15 UTC

lhcathome-dev et LHC@home won't be merged. It is similar to Seti et Seti beta.
Beta applications from WCG won't be merged.
ID: 5359 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2

Message boards : Number crunching : More merger proposals

©2024 Sébastien