Appeal for Help with Sibley Project

[pjones235]

I haven't mirrored any links yet. People aren't deleting the links..

[coulonnus]

I haven't mirrored any links yet. People aren't deleting the links..

But you could delete these links people didn't delete, when you find one. I've deleted a number of links that I found in this situation.

[icactus]

I haven't mirrored any links yet. People aren't deleting the links..

Before 2010 or whenever it was that the {{SibleyScan|1802/xxx}} template was created, files were not mirrored with the word "Sibley" in the file name, so they were impossible to track after being uploaded. But that really only affected files with Sibley numbers below 15,000 or so. Now that we have gotten through most of those, the remaining links have been filtered several different ways by different scripts I've run to remove already uploaded files as best as I can. I have a script that I run every day or so that adds the latest Sibley Links to the link list and it automatically searches the uploaded files log to make sure the file hasn't already been uploaded. We can do that because every file mirrored from Sibley contains the sibley number in the file name when it is added to the IMSLP servers. I started running that around the time that 18,000 was the latest Sibley number so if you want a less frustrating experience I would work on the files above 18,000.

Now that we have the Link List most people are deleting the links as they go but some people are too fast (Like Massenetique) and follow the RSS feed for Sibley and upload files the day they are posted by Sibley and before my script transfers them to the link list. Even in this case, the script still catches them because it checks the IMSLP upload log before adding the link to the list and sees that the file has already been uploaded.

So once we get above 18,000 there should be zero instances of files on the link list having already been mirrored unless Sibley rescans the same file twice (which has happened) or duplicates a file we already have from somewhere else.

And thank you in advance for your extra time helping!

[icactus]

FYI, there's no need to notate sibley links that are books or non-music files. If it isn't a score or a method or Libretto I'd just delete it.

[pjones235]

Much appreciated!

[icactus]

Good news! The first 19,000 files are nearly done! There's about 30-50 files in each of the groups from 12,000-19,000 that need to be uploaded and we can cross off pages 3 and 4 of the link list. We've had some good help identifying the files but now we really just need some people to do actual file transfers and we'll be in good shape!

PLEASE AND THANK YOU!

[icactus]

With the WIMA project finally in the rear-view mirror I'm hoping we can build a similar effort around getting up to date with the Sibley project. As of now there are only 1700 files that have yet to be labeled or uploaded and then we will be up to date. As the summer ends it will take more people to get caught up as more of us have less time to devote to this so we could use a lot more people than we currently have.

Once we are caught up to date there is a separate task of splitting the many collections that have been flagged in the link list and uploading them individually. But that can take the back seat to getting caught up for now.

Any WIMA uploaders care to join us?

[coulonnus]

I am still participating, but a "feature" is increasingly slowing down my work: each time I do a minor edit on a work page after I made a Sibley upload my browser remains stuck for seconds, showing http://pchnote.appspot.com/. Does Peachnote donate $.000001 to imslp each time imslp fetches something from this URL?

[imslp]

I will ask Vladimir about this issue, because I know as little about this as you. The Peachnote code is maintained solely by Vladimir and my only control over it is whether the Peachnote features are included on IMSLP at all. I will ask him to look into this matter.

[imslp]

I have received the following reply from Vladimir:

"I couldn't repeat the behavior myself, and I can't think of a reason why that happened. I am not tracking new submissions, so that could not have been the reason. I guess a thorough javascript tracking could help pinpointing the bug, but that would require some time to set up. If I'd get a more detailed information about the browser version, exact action sequence etc, in other words a repeatable description, that would help."

Would you be able to provide a more detailed description of the issue? Thanks!

[coulonnus]

I. If I'd get a more detailed information about the browser version, exact action sequence etc, in other words a repeatable description, that would help.

Sorry for the delay. Finally the Peachnote slowdown occurs only with one browser, Flock. So I'll use another one.

[coulonnus]

On the 16.000-18.999 list page all files, which are PD in Europe are now uploaded [...]

There are quite a few PD-worldwide scores in the 19.000-Present section.