Far too many typos—suggestions

Brian Wilson

I have been using Logos 4 for mac Scholar's addition for several weeks now. Seems that every time that I open a sermon or commentary, there are typos—some major. Just finished reading Spurgeon's sermon, "Confession of Sin—A Sermon with Seven Texts". There were 17 typos that I reported. Some of them were very obvious. Others I had to go to archive.org and look at the scanned version to discover what the original said. 

My internet connection is shaky. It has taken me over an hour to look up all of these typos and report them.

I have some suggestions:

1. Please provide an automatic log where users can see which typos they have reported. Then as the internet blinks on and off, the user knows whether the typo reports actually went through or not.

2. Please block the making of duplicate typo reports from the same user (where both the selection and the suggestion are the same.) (In three months from now if a typo that I have already reported, I probably won't remember reporting it and will submit another report.)

3. Please provide some form of incentive for submitting valid typo reports. Perhaps a small credit for each valid first typo report. I just submitted 17 typo reports. Some of them are probably the first time anyone has reported them. After a few months, I might have built up enough credit to purchase a new resource for my library. 

4. Provide some time frame for when we can expect that the corrections will be made and our resources updated.

5. Please provide an online database of scanned pages for users to access for the purpose of making typo reports. (or to put it another way, for the purpose of trying to figure out what in the world is going on with this sentence that is obviously incorrect.) Users could only access pages of those resources for which they have purchased. It would be optimal that when reporting a typo, there was an option that would open the browser to the scanned page of the book.

As we are about to be flooded with many more public domain books, something must be done to improve the quality.

DominicM

Yes, its so bad I have a highlighting pallete especially for the job.. and the markup on a shortcut.. just running a spell checker thru text isnt good enough...

and this one is even worse

how this failed even a basic spellcheck and then the proof read is beyond me...  

Mark Barnes

https://community.logos.com/discussion/comment/189166#Comment_189166

On another thread, Bob has said that most of the typos exist in much older material (and some third party material) which was OCR'd. Nowadays text is double-typed to ensure a higher level of accuracy.

But I agree with all your other comments. Rewarding users for reporting typos by giving us data that gives confidence that the reports are acted on in a timely manner is crucial in my view.

Dave Hooton

Brian Wilson said:

1. Please provide an automatic log where users can see which typos they have reported. Then as the internet blinks on and off, the user knows whether the typo reports actually went through or not.

You should get an email acknowledgement of your typo reports (one per day I think). Not quite what you were asking, but read this.

Brian Wilson said:

2. Please block the making of duplicate typo reports from the same user (where both the selection and the suggestion are the same.) (In three months from now if a typo that I have already reported, I probably won't remember reporting it and will submit another report.)

There is a strong possibility that your typo report is not unique, but would it make any difference to know that was the case? Checking the correctness or status of a possible typo will take more of your time for (IMHO) little return. Send and forget; let Logos validate your report. They will know that an old English spelling or word is appropriate; one reason that an automated Spell Checker is not the Quality Control solution that many users think. Logos' aim is to reproduce the original because they don't always have the freedom to make ad hoc alterations to the text they have been given! 

Dave Hooton

Brian Wilson said:

5. Please provide an online database of scanned pages for users to access for the purpose of making typo reports. (or to put it another way, for the purpose of trying to figure out what in the world is going on with this sentence that is obviously incorrect.) Users could only access pages of those resources for which they have purchased. It would be optimal that when reporting a typo, there was an option that would open the browser to the scanned page of the book.

The scanned pages should incorporate any corrections approved by Logos, for then you should be able to figure out what is happening and this would be useful beyond the purpose of making typo reports and wondering when the resource will be corrected.

nicky crane

https://community.logos.com/discussion/comment/189182#Comment_189182

I probably report an average of about 10 typos and metadata mistakes a day, mostly from USB Translator's handbook.  Very seldom do I find a typo that stops me understanding the meaning of the text.  It doesn't take a moment to report a typo, nor are we obliged to do it.  

I would appreciate the Logos3 facility for storing typos reported when offline and sending them when back online.

I have thought maybe a small credit per 10 or 100 original typos reported, but maybe this would overburden Logos staff further and add to the present inefficiency that so many of us, myself included, are unhappy about.  Or necessitate employing more staff, which would push up the price of resources for everyone.

DominicM

https://community.logos.com/discussion/comment/189177#Comment_189177

Mark Barnes said:

Nowadays text is double-typed to ensure a higher level of accuracy

But "Christians Daily Walk" was released in the last few weeks, so by that ruiling, it should have been caught..

nicky crane

https://community.logos.com/discussion/comment/189238#Comment_189238

DominicM said:

so by that ruiling, it should have been caught

We have typos even in the forums!  (I've done some myself, come to that....)

Mark Barnes

https://community.logos.com/discussion/comment/189166#Comment_189166

DominicM said:

and this one is even worse

how this failed even a basic spellcheck and then the proof read is beyond me...

Except it's not actually a typo. That's exactly what the original book says:

DominicM

https://community.logos.com/discussion/comment/189277#Comment_189277

ok - I stand corrected, just need an interpreter [:P]

Mark Barnes

https://community.logos.com/discussion/comment/189281#Comment_189281

DominicM said:

ok - I stand corrected, just need an interpreter

According to my COED, seizin is a variant spelling of seisin. Now I'm sure you know what seisin is [;)], but just in case it means possession, especially of land. That still didn't help me, because livery is something to do with horses where I come from. But apparently "livery of seisin" is " the ceremonial procedure at common law of conveying freehold land to a grantee".

The phrase "livery and seisen" is a favourite of Matthew Henry's.

(None of this I knew until ten minutes ago!)

DominicM

https://community.logos.com/discussion/comment/189291#Comment_189291

thanks, has also just done a library search for the phrase..

I also understood it as horses.. hence the confusion

Matthew C Jones

https://community.logos.com/discussion/comment/189295#Comment_189295

True typographical errors should be caught with double keying . It is the archaic (non-standardized) spellings that may slip through.That is why Noah Webster went on a quest to standardize spelling rules. I have found it very interesting to see an author spell a word differently within his own work. When Logos works with digital versions that were scanned from antique sources these must be updated.

fgh

https://community.logos.com/discussion/comment/189291#Comment_189291

Suggestion:

livery < leave  (cf delivery)

seizin < seize

Hence 'livery and seizin' = the act when one person legally gives up/leaves, and another legally takes/seizes, possession of something.

Or what do you think?

Brian Wilson

I would like to thank everyone for their replies. 

I read a lot of Spurgeon, Whitefield, and Matthew Henrey. Some examples where I didn't need the original:

Why cloth the proud = Why doth the proud

BAALAH = BALAAM

and it is A style = and it is a style

Some examples where I needed the original and was thankful for the scanned online copies (googlebooks and archive.org):

fight against trio = fight against his

indict-rubber = India-rubber

alas I for the sin = alas! for the sin

Job 6:20 = Job 7:20 (this was especially troubling as it involved a link ).

I have created a palette to mark-up the typos as suggested. Tested it on some text and now I can't unmark it!! Created a "normal text" highlight but it doesn't work. Also tried searching the marked text in open books for *, and after 20 minutes without finding my text, I gave up with that. I am bouncing around in my studies looking at different sources and it is not always convenient to report the typos, especially those that take a little research to decide what the original is supposed to be. 

I realize there is an issue with archaic spelling. When I had a question, I looked up the original online (again thanks to googlebooks and archive.org). But even still there were 17 typos in the one sermon. For the benefit to the logos community and to improve the quality of this wonderful software, some of my suggestions or something similar should be implemented. Spending an hour sending in the reports using an unreliable internet connection, not knowing if they were really logged or not, "but they probably were, hopefully" really discourages people from reporting and ultimately reduces the quality of the software.

Some of my suggestions are coding issues. If the server logged typo reports by user, but didn't pass on exact duplicates, (I suppose "exact duplicated" should include the comments box) then 1. the user has the satisfaction of knowing that he has actually reported something and 2. the logos staff isn't flooded with the same request coming in. (This would hopefully reduce work load and increase the speed for which the resources are updated.)

Which brings up the question...are the resources updated? I have invested quite heavily and am really struggling with the 10th commandment on several older collections that would end up costing me over $1,000. But the incentive to purchase is really missing if there is no follow up on these typo issues, and if we are just reporting into the wind.

Mark Barnes

https://community.logos.com/discussion/comment/189521#Comment_189521

Brian Wilson said:

If the server logged typo reports by user, but didn't pass on exact duplicates, (I suppose "exact duplicated" should include the comments box) then 1. the user has the satisfaction of knowing that he has actually reported something and 2. the logos staff isn't flooded with the same request coming in. (This would hopefully reduce work load and increase the speed for which the resources are updated.)

Actually it's best if the server does record duplicates (so long as they're marked as such). Knowing 20 users all reported the same typo (a) gives confidence that the report is correct, and (b) that the resource is heavily used and a correction worth prioritising.

Brian Wilson said:

Which brings up the question...are the resources updated?

Bob Pritchett (Logos CEO) recently said this:

"While some work has been done on typo corrections, we have run into a
significant delay because the system we use to store the typo reports
was not designed to scale up to the massive number of resources and
users that we are now trying to accommodate. We are in the process of
evaluating some different solutions to this problem, but it will be
awhile before we will likely have a better system in place."

I take that to mean that typo reports are being stored, 'old' typo reports will be merged into whatever the new system happens to be. So it's worth reporting typos, though we won't see the benefit for a while.

It's also worth pointing out again that new resources have (generally speaking) far fewer typos that old resources.