I just printed out the CastingWords giude.

Discussion in 'CastingWords' started by ewd76, Jan 28, 2013.

  1. ewd76

    ewd76 User

    Joined:
    Oct 8, 2012
    Messages:
    1,133
    Likes Received:
    0
    It seems easier than scrolling up and down on my screen. I'll take a couple of weeks to read and digest it before I try one.
    :twitch:
     
  2. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    Good luck!

    Apparently, there's been a lot of positive feedback about the Guide to CW from Turkers, particularly their top transcribers. I've gotten private messages and blog comments from folks who really like it, too. So, hopefully you'll find it well-organized and logical...or at least comprehensive. :)
     
  3. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    Not nuts about it so far.

    * The additional rules for speaker labeling should apply to editors, not transcribers. That's not our job. We transcribe the audio, the editor makes the transcript more useful for the customer. If CW wants us to do both jobs, they should pay us for both jobs. Let's be realistic, for the horrible pay we're getting, CW has no right to demand something like that.

    * Most of those rules are nonsensical on top of that. There should be one way to label an audience speaker and one way only. Exceptions slow everything down and add confusion.

    * Bullet points with no vertical spacing is an eyesore.

    * No table of contents.

    * Most or all instances of "Do this" should be replaced by a correct example and an incorrect example of what was just explained, so that there's context.

    * Still has important information located in other documents. Rare situations and the sample transcript should be integrated into the style guide. An ideal style guide should actually make the sample transcript redundant.

    * The sample transcript is useless. It's missing 90% of the rules from the guideline because it's so short. It's also not representative of the audio we normally see. It should have things like inaudible words, pauses, odd sounds that need to be transcribed (sounds we should transcribe and sound we shouldn't), and the audio it came from should accompany it.

    * Still missing information, such as guidance on handling quotations.

    * Still wrongly dictates the use double dashes and ellipsis. See http://en.****pedia.org/****/Em_dash#Em_dash. If the guide is telling us to do something that is wrong after lecturing transcribers about using good grammar and punctuation, that's a huge problem.

    * There are many unnecessarily confusing sections like the very first one. First it says "No special instructions can ever counteract certain CW requirements", which is followed by "Aside from the exceptions above, when your special instructions contradict the style here, follow the special instructions."

    That is an impossible contradiction. It should have said something along the lines of:

    "Special instructions on a CastingWords transcription HIT may override any rule in this document, except for the following: ..."

    * The first five bullet points after that are aimless. They should exist within other sections.

    * The last bullet point of the five is invalid. It says to use word wrap, which is a display function whose behavior will vary from program to program. It is not a formatting function. If you turn word wrap off in UltraEdit for example, it'll create a scroll bar in your document and keep everything you type on a single line until you hit the enter key. If you activate word wrap, it will get rid of the scroll bar and visually break long strings of text across multiple lines on its own. But the text itself remains the same in both modes. Copying a long paragraph to the clipboard in either mode will result in identical text: one long string.

    Shouldn't talk about program functions, should say "Keep each paragraph on its own line; don't spread it across multiple lines by hitting the enter key."

    * Repeats the invalid directive later on: "Word wrap must be turned on" and "Do not use software that does not have word wrap". Not only is this wrong, it's also redundant.

    * Makes no more clear the right way to handle paragraph length. Still sticks with the simplistic "no more than 400 characters". For many months I was punished for keeping paragraphs under 400 characters, but also keeping them nearly 400 characters, because the guide didn't say anything more specific than the maximum character count. Not one editor, grader, or CW employee told me what I was doing was wrong, but once I figured it out for myself, my grades began to improve.

    Should say "No more than two or three sentences, and certainly no more than 400 characters, whichever is shorter."

    And ideally, it should be brutally honest and say: "Visual pleasure should trump the transcript making sense", because that's what CW really wants.

    I could go on like this for quite a while. This new guide is a step up from the last one, but it still needs a lot of work. It's going to slow me down significantly for a while and for no good reason. It'll take me as long to memorize the new pointless speaker labeling requirements as it will to read, re-read, and read again all the confusing passages so that I get the rule while ignoring all the contradictions and poor structure.

    Hate to be cold about it, but this new guide hurts as much as it helps.

    On another note: I'm so sick and tired of this forum not working. One of the reasons I'll pop up for a day or two and then vanish for months at a time is because I'm sick to death of having to clear my cookies before a login will take, only to try to post something and see this: http://i.imgbox.com/adbyx2aJ.png

    No other vbulletin site online does this to me. Just this one. And it drives me away.
     
  4. Andy

    Andy Administrator
    Staff Member

    Joined:
    Jan 25, 2009
    Messages:
    2,218
    Likes Received:
    21
    What browser / OS are you using? I have never encountered that problem.
     
  5. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    Firefox 19.0 and every version that proceeded it for as long as I've been using this site (auto-update); Windows 7 Ultimate 64bit.

    No proxy.

    I have "Accept cookies from sites" and "Accept third-party cookies" enabled, "Keep until" is set to "I close Firefox", and mturkforum.com is an "Allow" exception (like gmail.com, amazon.com, mturk.com, etc) so that cookies may be persistent. I also use NoScript with allow exceptions for mturkforum.com.
     
  6. Andy

    Andy Administrator
    Staff Member

    Joined:
    Jan 25, 2009
    Messages:
    2,218
    Likes Received:
    21
    You also tick the "Remember me" box upon login? Our cookie settings via the vB ACP are set to the standard/default settings.
     
  7. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    I've done that before and it makes no difference. There are times when I haven't been here for months, and I'll put in my login info, get the "Welcome, you're getting logged in blah blah" interstitial screen, then return to the forum listing not logged in. And I have to delete all my cookies for this domain in order to get logged in even just once. Then after browsing around, sometimes just one page later, I'm not logged it again.

    Beats me, too. This doesn't happen on any other vB site for me. Any site at all. Who knows.
     
  8. nobody

    nobody User

    Joined:
    May 23, 2012
    Messages:
    317
    Likes Received:
    0
    Is anyone having trouble with the [sp] and [?] before the word?

    It's slowing me down. When I hear a word I'm not sure about, I can't just mark it and keep typing. I have to either mouse back or tab back. Some transcripts have a lot of words that you can't find, so this could be a huge problem. Do we know why this was changed?
     
  9. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    I don't spend any time trying to figure out spelling or inaudibles while transcribing precisely because it slows you down. I don't stop typing unless I've fallen behind the speakers. If I can manage an [xx] without having to stop the audio, I will. (Keeping "[xx]" in your clipboard so all you have to do is hit control-v may help). If I can manage a guess, like [xx word], I will. If I do have to stop audio playback, I have Express Scribe set to back up a few seconds after every stop which can really speed things up, although it doesn't seem to work every single time. Even a simple xx without brackets should be easy to mark the area so that you can continue.

    Once I'm done transcribing, the first thing I'll do is fix all the misspellings which are easy to see in any word processor. The second thing I'll do is rewind the audio and listen to it while reading the transcript to catch missing/wrong words, remove filler words which I'll purposefully leave in (I usually type verbatim, just for rhythm), break up paragraphs, and of course try to nail down inaudibles. I find that doing that step last results in better transcripts because the more you hear somebody's voice (not just accent, but speaking style) the easier you can pick out hard to hear things.

    It also helps to pre-screen the audio. I like to use finger counts. If I hear something that I don't understand, I put up one finger the way a child is taught how to count. I don't have any set numbers or a rule of thumb here, but if I don't feel good about the number of fingers I have up compared to the length of the audio that I've listened to, I won't do it. It's not worth the money to do it. Even if it takes you 10-15 minutes to find a piece of audio that only has one or no inaudibles, you'll make up that time by then doing a search for that job ID and doing *all* of those audio pieces for that job until they are gone.

    Look at this from the CastingWords HIT list:

    George Herming: 145071 (avg rwrd+bns: $0.68) [04:00 mmss] $0.49

    This is the same HIT on mTurk:

    Audio Transcription: Easy Chicago Easy #145071 (145% premium) (avg rwrd+bns: $0.98) [04:00 mmss]

    If you like that piece of audio, if it's quick and easy, do a search on mTurk for that number: 145071

    It'll give you every piece of audio (out of order) from the customer that's still available. Find a HIT that you're comfortable with and then exhaust all the HITs from that customer. I did that for 1.5 hours this morning and ended up making about $6.44. It's not what some transcribers on here can do (5/6 were auto-approved grade 8s and a single manually approved grade 9) because they are just plain faster (or less picky) than I am. But that'll help. I made more money this morning doing 6 HITs all from the same source than I ever would have made doing 6 HITs from 6 different sources, having to learn at least two new voices every 8-11 minutes, having to learn to mentally block out six different background noises.

    Short version: you have to learn to do that. Type xx even without the brackets, type anything that will allow you to keep going. Then go back and fix it later.
     
  10. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    Hey, pwt,

    Thanks for speaking your mind and offering such a detailed critique! Since I was the lead writer on the revision, I thought I'd offer a detailed response. I'm going to have to continue it in a second post; the system is telling me it's too long for one.

    Hm. It is a transcriber's job to label the speakers in a way that is easy to follow. That's not just a CW thing; it's an industrywide transcription thing. Editors don't have the job of labeling speakers in the first place, although they do need to make sure that it is done consistently across a transcript.

    Not sure where you're getting the crappy-pay thing. I think it's common to assume that because a company is on Turk, they must not be paying well, but the truth is, top pay at CW is what is considered very good transcription company pay off of Turk, too. :) That's based on the overall pay standards for general (not legal or medical, so we're comparing apples to apples) transcription.

    Honestly, I was surprised to find that out, too, when I did the research, but it's true. Plenty of transcription companies in the off-Turk world pay much worse, so suggesting that CW is at the bottom of the barrel isn't factual; it's not true on Turk, and it's not true off Turk, either.

    Agreed on both counts, but...you'd have to take these points up with CW directly. I also did pass your feedback on to them (without a name, of course). The labeling exceptions were as pared down as I could negotiate them to be; CW has final say on that. But I've mentioned on my blog before that I'd love it if the formatting were fixed; it's still not quite formatted the way I submitted it! Indents are missing, too, and vertical spacing throughout is inconsistent.

    It would be pretty unusual for a transcription style guide to have a TOC. But never a bad idea, so I've passed this on, too!

    There are way too many incorrect examples possible, which is why they were left out. The important thing seems to be how to do it correctly, so that's what's included. The fear was that if someone did something incorrectly but in a different way than was shown to be incorrect, they'd start complaining. After all, the person didn't do it "the incorrect way," so why was a transcript downgraded? (Unfortunately, based on some of the questions that CW describes that they get, this seemed like a realistic possibility.)

    Also, including all conceivable incorrect examples would definitely have made the style guide way longer, and I don't believe it would have added good enough information to make the extra length worth it.

    Yes, it does have information at other links. There was no way I was willing to incorporate the short rare situations guide or the short sample transcript into the main guide, though. Too much overwhelming detail in one place, and the reason for splitting them out is that they're not crucial in most cases. If, as you say, the sample transcript is redundant or useless (as you mentioned below), that was even more reason to leave it as a separate reference.

    I'm surprised to hear this, though I'd love to hear more about this from others, too. The sample transcript actually includes most of the rules from the guideline, including instances of inaudible words, pauses, and transcribed sounds. The idea was to make it short yet include a good idea of the format and approach to a transcript, while illustrating the correct use of the style points from the guide.

    Having a sample audio to go with...well, that isn't the goal of most transcriber style guides. In fact, people charge significant money for courses in how to do transcription that provide resources like that. So I don't think it's unreasonable for CW not to have an audio along with their sample. Many companies don't offer any sample at all, just their guide---and in most cases, that guide will be much less comprehensive than CW's is now.

    What would you need included? I know we've talked about quotes on this forum before...anything specific?

    Yep. I addressed this issue in this thread. No matter what, the client is still the final arbiter of what their style is, and with any company, their desired style supersedes normal grammar.

    Good rewording, thanks! I've passed it on.

    I personally think they make the ground rules clear for those new to CW and are clearly indicated under some explanatory text. But where would you like those placed?
     
    #10 naturegirl, Feb 12, 2013
    Last edited by a moderator: Feb 12, 2013
  11. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    Sorry you don't like the whole "word-wrap" terminology. However, the majority of transcribers do use programs where word-wrap is given as an option, and it's actually more confusing for many of them to say, "keep each paragraph on its own line," because next we'd have people trying to adjust their margins to get paragraphs to be all one line. So I think for general consumption, I stand by the reference to "word wrap."

    Also, the reason it says both "Word wrap must be turned on" and "Do not use software that does not have word wrap" is because some transcribers have wondered whether if they can't turn word wrap on (because it doesn't exist), they can use that software. The answer is no, it's not recommended.

    Thanks for the rewording suggestion! Passed on.

    LOL, you know that can't go in the style guide, because...it's not CW's official stance. Style guides are as much a reflection of policy as of...style.

    Also, I have not found it be true that visual pleasure trumps sense, although both are important. Any time (not often) that I've felt an editor or grader made a change for prettiness that didn't make any sense, I've gotten it reversed. You are right; some graders and editors only focus on whether something looks pretty. But CW does not think this is correct, any more than you do. If it happens, please let them know.

    No worries! Some people love it (the original release announcement thread was here), some don't. CW has said they are really surprised that people weren't regularly reviewing the old style guide anyway, to make sure they kept up with changes. So that seems to be their expectation.

    They don't consider the speaker labeling new (besides the audience stuff), to be honest. That was the way they wanted it done; they just never put it in print before.

    By the way, anyone would be welcome to ask CW if they can do the next version of the style guide instead. They happened to choose me, but I don't see any reason why they wouldn't let someone else do it.
     
  12. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    Hey, nobody,

    Sorry it's slowing you down. Yes, I can tell you why it was changed. Officially, it was always CW style, but it wasn't enforced (they were actually surprised to look back and realize it was supposed to be their style, because they had gotten used to it the other way, too).

    But, the reason it was changed is because the editor must replace the [sp] and [?], which never should be in the transcript delivered to the client. The point of those tags is actually to alert editors to an iffy spot. So, it made a lot more sense for editors to get a heads-up that the word(s) they were about to see was something the transcriber was unsure about, rather than about the word(s) they just saw.

    I do as pwt does and just type "xx" when I'm not sure about something. I either go back and replace it with the tag only or with the attempt and the tag. It sounds like you have a different approach, though, and I can see why that would require more adjustment.
     
  13. ewd76

    ewd76 User

    Joined:
    Oct 8, 2012
    Messages:
    1,133
    Likes Received:
    0
    That's my problem. It's too comprehensive because there are too many rules. That's why I think I will just stick to Claritrans if I'm going to transcribe. Even though they pay much less. It would be nice if they could have hits that were several minutes longer and paid more. But that's a discussion for a different time and place.
     
  14. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    This one social news site I've used for the past six years recently rolled out a new design that limits comments to 2000 characters, which is about 350-370 words. I don't think some people understand what they are doing. You don't write software and invite people to use it to talk, and then place arbitrary and unreasonable limits on how much they can say.

    I just don't understand that thought process. Sometimes there are things that can't be said in 140 characters (Twitter) or 354 words (Newsvine).

    I feel your pain.

    Thanks for the chat, here's my bit...

    Man 1 and Woman 1 is easy, but Interviewer and Interviewee is not. It clarifies roles but slows you down (while reading) because the words are so similar. If our job is to make the transcript easier to read, then we're intentionally failing.

    My main complaint though are the special rules involving audience members, where labeling changes depending on the number of people speaking and the type of audience. To me, that's just silly.

    I realize that some people are faster at this than I am, but not that much faster. If I can't make minimum wage doing this then there are few that can. For the work CW is asking for, yeah, the pay is crappy. It doesn't matter to me that all pay is crappy on mTurk, that doesn't strip me of the right to point it out. Like it or not, a more complex and thorough style guide with no increase in pay is a de facto decrease in pay because it's more work. (It's the same situation I feared would occur with Netflix. No mail delivery on Saturday means 48 fewer shipping days per year, which means fewer DVDs per year, which is a de facto price increase even if prices stay the same.)

    I don't know what the industry standard is for salaried transcription, but it has to be above $7/hr or it'd be illegal. I can type upwards of 140 words per minute and I struggle to make more than $3/hr working for CW when you include total work time (not just a the time it takes to transcribe).

    Did you you compare CW to other companies that farm out work to contractors, or companies that keep people on salary? I don't think that first comparison is anything close to meaningful.

    Thanks, but you can stick my name on anything you want in the future. I don't have a privacy issue with this. I like accountability.

    Yeah, if you try to list every single one. You don't need to do that. And maybe examples aren't necessary for every item. They could be really helpful with currency and breaking up run-on sentences, though.

    It should be. I used to do management for a chat network many years ago and we had a constant problem finding volunteers to do tech support (very simple tech support) that could pass the tests we created (simple tests, too). My analysis concluded that the problem was people weren't prepared for the test, and that expecting people to pass a test they didn't study for -- weren't taught for -- was practically guaranteeing failure.

    Although I ended up leaving before being able to draft a solution, my line of thought was that if you want people to do a job well, then you have to train them for it.

    I don't suggest CW to start training people, but providing sample audio to go along with a sample transcript will go a very long way towards teaching people how to handle the audio *for a fact* without having to guess or make inferences. It would be far more valuable to hand transcribers the audio and transcript, and say "this is what it sounds like, and this is what that should look like", rather than just "this is what something should look like" generically.

    Anything. I spent half an hour one evening trying to find out how to quote more than one paragraph of text and found several legitimate ways to do it. Are we supposed to quote thoughts? Blah blah blah and here I am thinking, "This guy's crazy!"

    Etc.

    I can't really answer that without rewriting it myself.

    I get what you're saying, but you're missing the point. It's not terminology, it's wrong. Flat out wrong. Word wrap is a display function, not a formatting function. The guide is telling people to do something that will have no effect. Word wrap is not the same thing as a line break.

    I can dream.
     
  15. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    I wish you luck, ewd, as always!

    Thank you!

    You're welcome for the chat. :) It's always nice to see you.

    Hm...the "Interviewer" and "Interviewee" thing actually isn't new. It just was never official, but it's always been their preference.

    SpeechInk does get around that in certain types of audio by using Q and A to designate the two parties. I doubt that CW will be willing to change to that, though. They didn't want to change to "Respondent" from "Interviewee," for example.


    I think you're missing my point: I was saying that people like to assume that because many payscales on Turk are crappy, CW's must be, too. But...it's not. Not compared to the transcription industry as a whole.

    You're complaining about level of work, but what CW asks is a standard level of work throughout the transcription industry.

    Hm, the style guide is more comprehensive now because everything they wanted people to do but weren't specifying you should do to get perfect scores is in it. I think that's an advantage to transcribers in the long run, which is why I was happy to join their project to redo it. (You've said already that you were often puzzled as to why you weren't getting the top scores in the years you've transcribed for CW.) And...style guide changes are a given in the transcription industry. You always have the choice to work elsewhere instead. That's the thing about being a contractor.

    Well...I moved up to making $12 and even $14 an hour with CW transcription in less than 6 months (I was at $10/hour within 6 weeks, and I worked very, very part time for most of the 6 months), so I'm pretty puzzled by your $3/hour figure. Although as I said in this blog post last year, typing speed is the least important skill in a transcriber's skillset. I don't type more than 90 wpm.

    I'm also not sure why you would think comparisons of CW to salaried positions would be fair. It's not a salaried position, so that'd be comparing apples to oranges. Most general transcription work in the industry is hired out to contractors. So again, CW's top pay is right in line with what are considered good contractor rates in the transcription industry as a whole.

    Gotcha! Passed on.

    You might try mentioning this to them directly; I think you can explain your idea better than I can. I bet they'll still say they're not in the business of doing so, especially now that they're getting more exposure to off-Turk and experienced contractors with the Workshop. I like your imagination and idealism, though!

    OK. Sounds like something they should put in the Rare Situations guide. I'll mention it to them too.

    You might consider asking them if you can do the next version!

    OK. You might try presenting this case to them directly, too. I suspect they'd have to train their transcribers to appreciate this distinction, too, and that seems like a big battle to fight, especially for a matter that doesn't officially have much to do with transcription, you know?

    Always. :)
     
    #15 naturegirl, Feb 13, 2013
    Last edited by a moderator: Feb 13, 2013
  16. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    I'm going to kind of drop the Style Guide discussion for a minute so we can focus on this, because this is important to me:

    Not nearly as puzzled as I am. The best guess I can come up with is that you're doing lots of (or even mostly) expedited HITs. Because the math doesn't add up for me. I did some HITs a few days ago, all from the same source because they were good quality, and my average for a two minute clip was 8.5 minutes of work, each paying $1.08. That's $7.62/hr.

    At your hourly rate, you'd have had to transcribe those two minute clips in real time without stopping even once, and then only do a single playback to check for errors in real time, also without stopping once, without spending even 5 seconds to move from one job to another and with no time at all for researching words -- and do that for an hour straight.

    If you can do that (which would require typing in excess of 240 words per minute), hats off to you.

    Maybe I'm a fool and I'm doing something wrong, but I just don't see how anyone can make much more than I do. I don't recall seeing anything in your book that would explain the speed difference. I would obviously love to understand this performance gap.

    Do you use a foot peddle? I know people say they can help, but I've gotten pretty fast with hot keys. But if that's the missing link... I can't see it being anything else at this point. If I can type almost 35% faster than you, how in the world am I making 75% less? I'm amazed. It's not like I'm getting G6 and G7s here. I get auto-approved at 8 and lately most that aren't auto-approved are 9s.

    Maybe I need to read your book again..
     
  17. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    I think your speed sounds great!

    I'm pretty sure, though, that there are several other CW transcribers with some experience under their belts who'd be puzzled that you would regularly be making as little as you've said. But as you're saying here, you made almost $8 an hour the other day, which is a far cry from $3 an hour. So...you know it's possible to do quite a bit better. There's just no reason a conscientious and grammar-conscious transcriber like you should be earning less than half minimum wage.

    It is all in the book, yes.:) I'll PM you, also.
     
  18. pwt

    pwt User

    Joined:
    Apr 28, 2012
    Messages:
    191
    Likes Received:
    0
    I keep meaning to make a distinction. I use two measurements of time, one is a stopwatch on my phone that starts when I start transcribing and stops when I'm pasting the transcript into the mTurk form. That's the ~$7/hr rate. The other measurement is a notation of the time when I start working, and again when I stop. That's how I end up closer to $3-4/hr.

    And that's even when I've found a good customer and am just blowing through one HIT after another from the same source.

    I really don't know how I'm losing that much time in between HITs. It's not like I get up and make a sandwich or go take a shower. I was looking at this last night (remember I'm a stats wonk) and found that I spent 117 minutes transcribing but 164 minutes working. That's either $5.70/hr or $4.10. Neither is anything close to what you can do.

    It's frustrating.
     
  19. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    I responded to this higher up in the thread, but...it's now changed again so that these tags go after the word, and this way no one has to back up while typing. Sounds like that was easier for transcribers, and editors will just have to skip their eyes back. :)

    Here's the thread I started about that change.
     
  20. naturegirl

    naturegirl User

    Joined:
    Aug 30, 2012
    Messages:
    701
    Likes Received:
    0
    Hm. For that part, I'm just not sure what to tell you. But...it's not really fair in any contracting situation to expect the pay to cover the time you aren't doing the work they're paying for, so that seems like it is deflating your pay artificially.

    In any case, I've written to you, so I hope our conversation there will help!
     

Share This Page