Crowdsourcing the linking of resources?

Page 1 of 1 (15 items)
This post has 14 Replies | 2 Followers

Posts 174
Derek Browning | Forum Activity | Posted: Sat, Dec 2 2017 2:23 PM

I've ran across several scenarios in my library where a resource is not linked to other resources in my library.  I recognize that the FaithLife algorithms that catch these scenarios are imperfect, and it would be extremely costly to have every resource perfectly linked (especially when some earlier published resources have been added after a more recently published resource that quotes it).

I wonder if there's an option to turn some of this over to the community to do the work of linking so that someone at FL would only have to approve the work.  Over time a smaller group of vetted individuals could even do the approvals?

I'd be happy to contribute as I use it and find these scenarios.

Thoughts?

Thanks,

Derek

Posts 27451
Forum MVP
JT (alabama24) | Forum Activity | Replied: Sat, Dec 2 2017 3:23 PM

Derek Browning:
Thoughts?

This is a common suggestion, but one which doesn't seem to have much momentum with FL. 

OSX & iOS | Logs |  Install

Posts 1005
EastTN | Forum Activity | Replied: Sat, Dec 2 2017 5:02 PM

alabama24:

Derek Browning:
Thoughts?

This is a common suggestion, but one which doesn't seem to have much momentum with FL. 

But it's one I believe has real merit.  I imagine it working much like reporting typos.  FaithLife staff would retain control over actually updating resources, but most of the heavy lifting would be done by users.

Posts 1084
Sean | Forum Activity | Replied: Sat, Dec 2 2017 5:58 PM

EastTN:

alabama24:

Derek Browning:
Thoughts?

This is a common suggestion, but one which doesn't seem to have much momentum with FL. 

But it's one I believe has real merit.  I imagine it working much like reporting typos.  FaithLife staff would retain control over actually updating resources, but most of the heavy lifting would be done by users.

It's been brought up before, and while not a bad suggestion for solving a real problem, why should users do "heavy lifting" for something they've already paid* FaithLife to do?

*Yes, I know, "old resources" "can't expect perfection" "you think they should do this in perpetuity" etc. etc. The level and quality of linking is very inconsistent across the catalog, and even some recent resources lack links to older ones for which there is no justifiable excuse.

 Logos Now Subscriber -- 22/2/2018

Posts 1347
Ben | Forum Activity | Replied: Sat, Dec 2 2017 6:00 PM

I don't think I've ever seen a successful crowdsourcing here. Floated one myself, and it went nowhere. 

"The whole modern world has divided itself into Conservatives and Progressives. The business of Progressives is to go on making mistakes. The business of Conservatives is to prevent mistakes from being corrected."- G.K. Chesterton

Posts 174
Derek Browning | Forum Activity | Replied: Sat, Dec 2 2017 6:04 PM

I would argue that to stay price competitive against other electronic and even paperback formats, it's not possible to have the level of detail and perfection. IE, an 80% book for a 10% premium is much more likely to sell than a 99% book at a 100% premium. So I'm OK with better prices and lower quality on the linking. The crowd sourcing option could be voluntary or people could earn credits for their work to apply to discounts.

Posts 28
Dave Palmer | Forum Activity | Replied: Sun, Dec 3 2017 7:47 AM

If I could choose one feature that I would add to Logos above all others this would be it.  Leveraging crowd sourcing to fill in the gaps of where Logos is lacking would take the product to another level.  

Posts 1005
EastTN | Forum Activity | Replied: Sun, Dec 3 2017 12:04 PM

Sean:

It's been brought up before, and while not a bad suggestion for solving a real problem, why should users do "heavy lifting" for something they've already paid* FaithLife to do?

Because it likely won't happen otherwise? Because if I have to chase down a reference anyway, I might as well take a moment to report it so you won't have to do the same thing 18 months later?  Because if we all do it we'll end up with better resources without any one of us having to invest in a huge amount of time?

Posts 2248
Jan Krohn | Forum Activity | Replied: Sun, Dec 3 2017 2:08 PM

alabama24:
This is a common suggestion, but one which doesn't seem to have much momentum with FL.

Dave Palmer:
If I could choose one feature that I would add to Logos above all others this would be it.

Maybe we need a UserVoice suggestion, and vote this up. But we need to put in a good process. So before I write it up, let me throw in a process for discussion. In the German sub-forum, a crowdsourcing project has just been stopped due to copyright issues (Schlatter's NT commentary), but the process seemed to me quite good:

  • The community provides the digital document
  • FL puts the product on Pre Pub
  • The community places bids until 100% funding
  • FL produces a resource from the document
  • On release, bidders pay their fees, and receive the same amount back in Logos credit (therefore receive the resource de facto for free).

Past IT Consultant. Past Mission Worker. Entrepreneur. Future Seminary Student.
Why Amazon sucks: Full background story of my legal dispute with the online giant

Posts 4885
DIsciple II | Forum Activity | Replied: Sun, Dec 3 2017 3:26 PM

EastTN:

Sean:

It's been brought up before, and while not a bad suggestion for solving a real problem, why should users do "heavy lifting" for something they've already paid* FaithLife to do?

Because it likely won't happen otherwise? Because if I have to chase down a reference anyway, I might as well take a moment to report it so you won't have to do the same thing 18 months later?  Because if we all do it we'll end up with better resources without any one of us having to invest in a huge amount of time?

Reporting it won't get it fixed, you can report it now through report typo. Users simply sending in reports of missing links won't work.  It links are to be updated efficiently they need to be done on a resource by resource basis otherwise FL staff would be going back to the same resource repeatedly and we was users would end up getting more resource downloads of the same resource multiple times instead of one download to fix them all. There are far too many resources in the catalogue for spot reporting to be a viable and efficient solution.

Linking is a fundamental concept nd selling point in the software the marketing of FL likes to highlight but reality  is those lofty goals can no longer be achieved.

FL needs to change their marketing to set realistic expectations of the Features and Benefits of the software.

User need to change their expectations to match reality of what can be achieved.

Reality is linking needs to be achieved at a price that ensures we as customers are not priced out of purchasing and FL is not priced out of the market.

Posts 943
Everett Headley | Forum Activity | Replied: Sun, Dec 3 2017 4:13 PM

We all bought this software on a major selling point of everything being hyperlink.  The truth is that it is not.  I have reported ad naseum links that need to be fixed to no avail.  Logos obviously does not consider this an issue worth spending resources on. 

The expectation of users should be no more than what is promised:  hyperlinks in resources we purchase.  

It is on Logos to fix links.  This isn't hard, a simple search when a new resource is being built would identify them.  I find it absurd and ridiculous when I purchase a new resource and it has links that are not working.

 Alas, I believe Doc is right in that future purchase need to be made not on the hyperlinking but on the features of the software.

Posts 617
Ryan | Forum Activity | Replied: Sun, Dec 3 2017 9:55 PM

Dave Palmer:

If I could choose one feature that I would add to Logos above all others this would be it. 

Yes

Posts 1005
EastTN | Forum Activity | Replied: Mon, Dec 4 2017 6:54 AM

Disciple of Christ (doc):

Reporting it won't get it fixed, you can report it now through report typo. Users simply sending in reports of missing links won't work.  It links are to be updated efficiently they need to be done on a resource by resource basis otherwise FL staff would be going back to the same resource repeatedly and we was users would end up getting more resource downloads of the same resource multiple times instead of one download to fix them all. There are far too many resources in the catalogue for spot reporting to be a viable and efficient solution.

No, reporting alone won't do it.  There are two basic possible crowdsourcing solutions: a) something like community notes, where the links users provide are directly made visible to other users, or b) something like reported typos, where the links are curated by FaithLife staff and distributed through updates to the resources.  For something as fundamental as linking, the second approach makes more sense to me.  If a link is "live" I want to be able to assume that it is correct.  That does have, as you note, the downside of triggering multiple updates of resources.  I would expect that FaithLife would - for efficiency's sake - collect submitted links and do periodic updates to manage the workload involved.

I guess you can do it now through a typo, but it isn't a great system for links.  Much better would be something where you clicked on the place where a link should be, then clicked on the location in the resource it should be linked to.

Disciple of Christ (doc):

Reality is linking needs to be achieved at a price that ensures we as customers are not priced out of purchasing and FL is not priced out of the market.

I agree completely.  That's why I think it would be helpful to do a bit of crowdsourcing - we could get more links for the same price, plus just a bit of sweat equity.

Posts 1084
Sean | Forum Activity | Replied: Mon, Dec 4 2017 7:50 AM

There are two issues about Logos I feel strongly negatively about. One is the lack of selective downloading (a common-sense feature discontinued with L4). The other is the poor quality of links—something that is continuously advertised as one of the reasons to buy Logos.

There are two problems with links. The first is older resources that have very poor linking; they were produced before many of the sources they cite made it into Logos. This is a more understandable situation as standards and expectations were different back then—except there is no indicator whatsoever on the product pages to the effect that “Hey, this is an older resource; the linking’s not so good.” The only way you can find out is by buying.

The second is problems with newer resources, which, for some reason or the other, don’t have all the links they can or should have. I’ve had a few come through in the last year or so that lack links to resources that have been in Logos for a longer period of time. Really, there’s no excuse for that.

I know FaithLife can’t spend the money needed to update all the old resources to perfection. I do think a reasonable balance can be achieved. For example, popular resources and ones that go on sale frequently (=some cash flow) like the IVP Bible Dictionaries should get upgraded with better links. FL does do this sometimes, but not enough.

As I stated earlier, the suggestion of crowdsourcing of linking is not entirely a bad response from users wanting to solve this problem. I understand where it comes from. However, it’s far from the best solution for several reasons:

FIRST, I think it is a VERY bad idea to send a message to a vendor that it’s okay for them to let quality slide here and there because users will make up the slack. Customers should not have to do for free what they’ve already paid FaithLife to do. (Offering credit for doing so is a different story.)

SECOND, the problem is not just a link missing here or there. In some resources there are two or ten or 50 or more reference even to a single work that are not linked. For users to have to go through all of these one by one, multiplied by however many is necessary for proper crowdsourcing accuracy, is a tremendous waste of manpower. I’m sure that FaithLife has its own internal tools for doing this sort of thing en masse that would be much more efficient than users randomly plodding around.

This leads to the biggest problem with this idea:

For this to happen, FL would have to develop, debug, integrate with Logos, deploy, and monitor a new tool and continuously take action on data collected by it in order for it to accomplish anything at all.

Knowing FaithLife as we do, what do you think the chances of that happening are?

 Logos Now Subscriber -- 22/2/2018

Posts 10043
Denise | Forum Activity | Replied: Mon, Dec 4 2017 8:17 AM

I like EastTn's idea on a community link. Even if Logos got bogged down with progressively more un-linked resources, users could quickly use their fellow study-er's help. If and how Logos might use the corrections later might be iffy ... not sure it'd save them time, since much is automated (or should be).

In my own Bible software, I do the same for me. Temporary fix. Code the note for later batch fixing.


Page 1 of 1 (15 items) | RSS