search

IALSA / IALSA-2015-Portland

Hub to accompany IALSA 2015 workshop at Portland, OR, Feb 22-25, 2015
GNU General Public License v2.0
2 stars 0 forks source link

Duplicates in EAS #106

Open andkov opened 9 years ago

andkov commented 9 years ago

@andreazammit , sorry to alert you, but complexity takes its tolls. Duplicates began to creep into your models. For example, "b1_female_aehplus_pulmonary_reasoning_pek_block.out" is stored twice: in "NEW models for PEK" and "PEK long Output Files" , the scripts have no way of knowing which one should be the "right one" and added to the model pool. Skimming your folder map I anticipate more of these duplicates, which break our scripts. Could you please check if you have removed all the models you are were not sure about, forgot to remove, etc? It would save me a ton of time on debugging.

andreazammit commented 9 years ago

@andkov done

ampiccinin commented 8 years ago

@andkov - I asked Andrea Z to run the new models. In a case where we want to be able to compare old with new (this may or may not have been one), do we need to save the old in a subdirectory outside of GitHub and then overwrite the old files in the original subdirectory? In the future, before asking someone to delete output, could you please ask first why they have "duplicates"? They may not actually be duplicates and we may want to re-name the old ones, or save them elsewhere, rather than delete them.

andkov commented 8 years ago

@ampiccinin , to make things easier for all of us, yes, if there is a reason to keep a copy, @andreazammit should keep a copy somewhere outside of GitHub. I am operating under assumption that EVERY model uploaded to GitHub has passes the quality control of the original author and is there because the author intended it to be there, not just because the file got forgotten to be deleted, or misplaced, or something else.

This exposes of the current weakness of the system, but we'll organize things better after revision. Great point about duplicates. I don't have an easy solution now.

ampiccinin commented 8 years ago

@andkov , I am fairly sure you will encounter this in many of the studies. Before you delete everything, can you give me some time to save meaningful duplicates somewhere on my PC as well? (not now as I’m about to move to other non-PC demands…)

From: Andriy V. Koval [mailto:notifications@github.com] Sent: Monday, December 7, 2015 6:30 AM To: IALSA/IALSA-2015-Portland IALSA-2015-Portland@noreply.github.com Cc: Andrea Piccinin piccinin@uvic.ca Subject: Re: [IALSA-2015-Portland] Duplicates in EAS (#106)

@ampiccininhttps://github.com/ampiccinin , to make things easier for all of us, yes, if there is a reason to keep a copy, @andreazammithttps://github.com/andreazammit should keep a copy somewhere outside of GitHub. I am operating under assumption that EVERY model uploaded to GitHub has passes the quality control of the original author and is there because the author intended it to be there, not just because the file got forgotten to be deleted, or misplaced, or something else.

This exposes of the current weakness of the system, but we'll organize things better after revision. Great point about duplicates. I don't have an easy solution now.

— Reply to this email directly or view it on GitHubhttps://github.com/IALSA/IALSA-2015-Portland/issues/106#issuecomment-162540145.

wibeasley commented 8 years ago

Once something is saved in a Git repository, it will be recoverable in the future. I'm not trying to discourage you from saving them if it makes sense. But unless Andre y or I make a huge mistake, we'll always be able to turn back the clock.

Sent from Outlook Mobile

On Mon, Dec 7, 2015 at 6:42 AM -0800, "Andrea Piccinin" notifications@github.com wrote:

@andkov , I am fairly sure you will encounter this in many of the studies. Before you delete everything, can you give me some time to save meaningful duplicates somewhere on my PC as well? (not now as I’m about to move to other non-PC demands…)

From: Andriy V. Koval [mailto:notifications@github.com] Sent: Monday, December 7, 2015 6:30 AM To: IALSA/IALSA-2015-Portland IALSA-2015-Portland@noreply.github.com Cc: Andrea Piccinin piccinin@uvic.ca Subject: Re: [IALSA-2015-Portland] Duplicates in EAS (#106)

@ampiccininhttps://github.com/ampiccinin , to make things easier for all of us, yes, if there is a reason to keep a copy, @andreazammithttps://github.com/andreazammit should keep a copy somewhere outside of GitHub. I am operating under assumption that EVERY model uploaded to GitHub has passes the quality control of the original author and is there because the author intended it to be there, not just because the file got forgotten to be deleted, or misplaced, or something else.

This exposes of the current weakness of the system, but we'll organize things better after revision. Great point about duplicates. I don't have an easy solution now.

— Reply to this email directly or view it on GitHubhttps://github.com/IALSA/IALSA-2015-Portland/issues/106#issuecomment-162540145.

Reply to this email directly or view it on GitHub: https://github.com/IALSA/IALSA-2015-Portland/issues/106#issuecomment-162543883

andreazammit commented 8 years ago

@ampiccinin @andkov I actually have copies of everything..... I tend to save in different places - let me know if you want me to dig anything up.

andkov commented 8 years ago

Problem solved. thanks, @andreazammit!