Chinese podcasts, videos and HSK tests

[trevelyan]

Hi Everyone,

After not-inconsiderable development and feedback from some members of this board,I'm pleased to announce the beta release of Popup Chinese. If you visit you'll see we produce daily lessons, videos, HSK tests and more from Beijing. We place a strong emphasis on open content, standard mandarin and objective ways to track progress:

http://popupchinese.com

I'm proud of our ability to offer a really customized accounts (full traditional support? yes!). Beyond customizing the lesson interface, what we're aiming for with Popup Chinese is an open platform for anyone to develop and publish their own custom learning materials. It's easy to create lessons: we automate almost all of the gruntwork involved in cranking out manually annotated texts, PDFs, vocabulary lists, tests and much more. If you can point and click, you can make a lesson.

We're looking for feedback on both the design aspects of the service, as well as the pedagogical approach we've taken in some of the early podcasts. For a limited time, we're also giving free premium access to early adopters as a way to say thanks. Especially given the help I've had here in the past, I wanted to post news about us on Roddy's forums first. To sign up use voucher number 2008AOYUN at the following address:

http://popupchinese.com/voucher

Hope to hear from you, and I hope you enjoy the site. If anyone has any questions about our plans or suggestions on features we should incorporate, I'd be happy to talk about them here as well.

--david

[Shadowdh]

Fantastic stuff... easy to navigate around and easy to use. I have only played with it for about 10 mins or so so far but will do a lot more in the next weeks as its looking to be a wonderful resource for study. The only thing I have found so far is the quiz/discussion pages on the HSK testing need to be either the other way around (so you dont get the discussion before the test) or you need a welcome page for that part. I like the way it shows the results too. Well done. Will post more as I go through it.

[davidj]

I can't work out the business model, which always makes me nervous about a web site.

[roddy]

There are paid subscription options once you're logged in and looking at lessons. I'd guess either you aren't logged in, or you've used the voucher, in which case a lot of what you're seeing is actually premium content.

Looks like a very promising new venture, high hopes for this one.

[renzhe]

It's certainly commendable and looks interesting.

But if people can create their own lessons (podcasts, videos, other content), how does that work? Do you have to subscribe to create content? Do you get access to other content for free if you create content yourself?

EDIT: Is there a way to pick your username? Right now, it seems like you're only allowed to use the username that is equal to your email address, and this is not a good option for me. There is no way to change your username once you log in either.

[davidj]

There are paid subscription options once you're logged in and looking at lessons. I'd guess either you aren't logged in, or you've used the voucher

I wasn't logged in. Generally I want the site to tell me that sort of thing before I'll tell them anything about myself. (I have similar problems with sites that will only tell me postage rates once I've created an account.)

[imron]

Is there a way to pick your username?

Go to your profile and click on your username. You can then change it.

Regarding the business model, all the vocab, quizzes, lesson creation functionality, PDF downloads seem to be for subscribers only. I would guess it follows the principles of give people enough free content to bring them to the site, and then charge for value-added extras.

[Luobot]

It’s a very promising looking site, and I wish you great success with it. The mp3’s have good sound quality, with notably clear voices (yours) and very good Mandarin pronunciation from the Chinese speakers, such as Echo. There were a few minor things that I noted as I toured the site. Hope you don’t mind.

Profile:

- Need acknowledgement when updating profile, such as, “Profile Updated”

- One button for updating all “Display Options”

- “Show Translations” under “Display Translations” – when checking and clicking update, the check mark disappears as if it hasn’t been saved.

Flashcards:

- When the user returns to the flashcards, it should remember the last mode selected rather than defaulting to simplified, or just use the “Character Preference” profile setting (but possibly I didn’t successfully update my profile? How would I know?)

- Also, when the user returns to the flashcards, it should randomize the order in which the flashcards are displayed (otherwise, it’s too easy to anticipate them when repeating the test).

- Needs sound files for each flashcard.

- And a flashcard test that plays the soundfile and has you identify the Chinese characters and/or the English translation.

- (Added): How about a flashcard export that also supports ZDT's native format.

HSK Beginner Test:

- Thought the PDF should have the answers on a separate last page rather than upside down, in order to make it easier to read.

- The text version appears to only have the first answer for 1-d.

Stories:

- I especially liked these. Most especially the Lu Xun story with mp3, pdf, text with mouseover, and text-only options. It speaks for itself.

Newbie lessons:

- I like the format and think these are great for newbies.

Well that’s it for this tour. 加油

Edited September 8, 2008 at 05:16 PM by Luobot
Added the Flashcard export feature request.

[jbradfor]

Wow, looks like a ChinesePod clone. Not that I'm saying it's a bad site, but wow, you pretty much copied everything. Even the "generative audio review that gets you speaking. A pioneering and mind-bending product." is a rip-off from ChinesePod. You didn't even have enough imagination to change the name!

On the plus side, I also really like the "short story" part and the "quiz" for each lesson. That, at least, is different than CP. And I like the web design, nice and clean.

Luobot above pretty much covered most of my suggestions. A couple more.

I selected traditional, but the short story showed up in simplified.

It would be nice if the "quiz" changed each time. Yes, that's more work, but all I mean is make something around 15-20 questions, and display 4 or 5 each time.

Your "discussion" part is really primitive. I've whined about this at CP too. No multimedia, no threading, no notification of updates. Just text. Looks like something from the 80's. A forum-type discussion area would be much more useful.

Big request: this is something that I've wanted for a while, and might really set you apart. Have the site keep track of all the words I know, and have it tell me which lessons are appropriate for me given my vocabulary. It should be easy for it to know how many words in each lesson I don't know, and once I know all the words, flag it as appropriate for me.

[Luobot]

Big request: this is something that I've wanted for a while, and might really set you apart. Have the site keep track of all the words I know, and have it tell me which lessons are appropriate for me given my vocabulary. It should be easy for it to know how many words in each lesson I don't know, and once I know all the words, flag it as appropriate for me.

If you are able to do that, then may as well put out the list of lessons in some order, such as:

The following lessons will add 1 new word to your vocabulary:

Lesson xxxx

Lesson xxxx

The following lessons will add 2 new words to your vocabulary:

Lesson xxxx

Lesson xxxx

The following lessons will add 3 new words to your vocabulary:

Lesson xxxx

Lesson xxxx

and so on.

[trevelyan]

Hi Guys,

Sorry for the delay in getting back to this -- my girlfriend's parents showed up for dinner an hour and a half early and then stayed late into the evening. Interesting stories about living through the Cultural Revolution in 黑龙江, but a bit of Internet withdrawal after a while. This sort of feedback is really important to us, so I do want to address all of the points and questions in turn.

First of all though, thanks Luobot, ShadowDH, and jbradfor for the usability and feature suggestions. You guys are great: this is exactly the sort of feedback we need and I've added all of your suggestions to our internal TODO list. A lot of the flashcard functionality has already been fixed (no audio yet, that will take some time). We're prioritizing usability issues now and I hope to get to many of the remaining issues shortly.

Am hoping to have some videos up-and-running shortly showing our (ugly) older logo, and some of the basic features. Will upload when they're ready.

On some of the more specific non-bug-repurt notes:

@davidj - yes, the voucher provides access to premium content. Our interstitial registration/signup page *should* only appear when someone tries to access features that require authentication or payment. I don't think we do a good enough job signaling what exactly is premium content right now. Registration requires a valid email address, but there's no reason people can't use anonymous ones.

@renzhe & imron - lesson creation is currently a free feature. We want to encourage creativity and sharing, even if people only use the service to generate a PDF or vocab list for private study or distribution to fellow students. In the long run we'd like to move towards a business model that allows us to compensate content creators if others use their materials (anyone interested in producing regular content should get in touch). In the meantime, all users have premium access to whatever lessons they create. If we get loads of users uploading the exact same MP3 and writing the exact same transcript we may change this in the future.

@Luobot - thanks. a lot of great suggestions. you can expect a lot of them to be implemented shortly.

@jbradfor - ChinesePod is a good company with some great people working for it. There's some overlap in the free content, but I think the differences are much greater than the similarities. Our underlying architecture, cost structure and pedagogical focus is very different and we're hoping to move in a different direction with our business model. And the Fix is the fix because it fixes pronunciation, tones and is a quick fix.

It's a really interesting idea to track/recommend lessons based on known vocabulary. I think the hard part is finding a way to ensure more advanced students aren't recommended every single Beginner lesson that comes out simply because the machine thinks they know 90% of the vocab. We're breaking down all of our texts and dialogues to the individual word level (even to the specific gloss) as part of lesson preparation, and so have a good basis for building more complex recommendation features moving forward.

It's a good question of how to do this. One of the real weaknesses with most A.I.-driven learning systems is the way they assume the computer has both the full attention of the student, and dominates his or her learning process. I think it would be great if we could come up with a more nuanced way of figuring out the difficulty level of specific content and recommending it to people on that basis. Suggestions are very welcome.

[trevelyan]

almost missed this one:

Your "discussion" part is really primitive. I've whined about this at CP too. No multimedia, no threading, no notification of updates. Just text. Looks like something from the 80's. A forum-type discussion area would be much more useful.

i agree that improvements to the commenting section are needed. We're not aiming to be a forum like Roddy's here, and I agree that there's no need to reinvent the wheel. Better comment management and WYSIWYG editing is a relative priority.

[trevelyan]

Hi Simon,

I wanted to reply to your comment from the other thread, but didn't want to sidetrack discussion there, so am posting here. Thanks for taking a look at Popup Chinese and giving us your feedback. If you have more detailed thoughts, I'd be very interested to hear them.

most of the material is behind a subscription wall and lacks a coherent scheme and good upper intermediate classes. I am not going to touch on the pods though the two I listened to seemed short..

Not including user-generated content, we will ultimately be publishing somewhere between 10 and 15 shows on a regular basis. We've just moved into beta and are still flushing out the archives. More difficult materials will be released shortly.

Our general approach to lesson creation is to build lessons around content rather than the other way around as seems prevalent. This means there is no set lesson length: if it takes 5 minutes to cover everything, the audio lesson will be 5 minutes. If it takes 20 minutes... grab a coffee.

Perceptive comment on the lack of an explicit scheme for leveling up by the way. There's a brief discussion about our approach to "difficulty levels" online here. The short version is that we don't believe in the mass hierarchical approach to learning. We like to think the learning track looks more like a 90 degree arch that opens up ahead of the student. It's easy to structure progressive material for beginners because the student knows so little. Things get much more difficult once you get past about 500 words. We will be using alternate approaches to provide assistance to students who want explicit structure.

This makes our site design and approach to learning - as far as I know - completely unique. I don't see much similarity with anything else out there..

I thought the whole point of discussing doing on Chinese-forums was to remedy these issues of having a subscription wall instead of paying for it with advertising or referrals.

As mentioned above, users get free premium access to any content they create, and are free to give away their own work if they wish. We are working towards providing content creators with ways to generate income from teaching.

Or I am not sure if trevelyan is willing to make it advertising based...?

The platform is already free to use and free of advertising. Anyone who develops content through the system can give it away as they wish. The choice always belongs to the user.

Thanks for taking the time to read this reply. We are hoping to get all of our usability issues and minor bugs worked out before leaving beta. If you have any questions, comments or suggestions they are always very welcome.

--dave

[Luobot]

We are working towards providing content creators with ways to generate income from teaching.

I realize these are still early days, but at the most basic level, one needs to know:

- How do you plan to compensate content creators?

- What rights do content creators keep and give away to the content they create?

- How will it work for content created by a group rather than by an individual?

At one time, listeners contributed their remixes to ChinesePod only to see them gradually disappear from sight. Visibility is everything. Without it, even Aric Queen couldn’t hope to succeed (and he moved on). So the issue of rights to the content is important, because if someone contributes content to PopupChinese, and it doesn’t get the attention that they think it deserves, then they should have the unhampered right to take it elsewhere.

The platform is already free to use and free of advertising.

But your current plan calls for PopupChinese to be a paid subscription site, which is fine, but there are already a lot of good, paid subscription sites out there. Our discussion here is about developing an alternative to that model. In short, the point isn’t to be free of advertising. To the contrary, the point is to use the advertising model in place of the paid subscription model in order to give all users free access to all content. This gets away from the class-based idea that there are “basic users” as opposed to “premium users” and then everyone else is a “freeloader.” I felt that this mindset poisoned the atmosphere at ChinesePod whenever the issue of user class and access to content came up.

So the question that Simon asked is whether you could be convinced to move to an advertising-based model -- and not just for the user contributed content, but for the entire site.

I like what you’ve done with PopupChinese, so far, and I’m looking forward to hearing your thoughts.

Edited September 11, 2008 at 03:04 PM by Luobot
typo

[trevelyan]

- How do you plan to compensate content creators?

Any workable model takes a portion of the funds from people who pay for content/access and allocate it to content producers. It's premature for us to talk about specific metrics at this point. If the incentive for anyone to create content at this point is to get paid, they should either get in touch so we can shortlist them for a trial, or wait several months.

We are interested in hearing from people with ideas on shows they'd be interested in running.

- What rights do content creators keep and give away to the content they create?

Close to Flickr. We require the rights to operate the service. Content creators can create, upload and distribute as they wish. We'll stick up the legalese before we slip out of beta and will be happy to talk about details and make changes as necessary.

At one time, listeners contributed their remixes only to see them gradually disappear from sight. Visibility is everything. Without it, even Aric Queen couldn’t hope to succeed (and he moved on). So the issue of rights to the content is important

I see what you mean. We see distribution/usage and visibility as two fundamentally different issues. The rights issue was touched on above. Our policy on visibility is to afford equal treatment to all publishers. Any user can subscribe to any other user.

We pick a set of default lessons for users who subscribe through the main site, as they've presumably come looking for our content. We will give content producers a way to get users they refer or study with automatically subscribed to their content.

Someone who subscribes entirely to non Popup content will have a different experience from someone who subscribes exclusively to our materials, but the site navigation, and lesson display features will work exactly the same. Publishers who expect visibility will have to produce the sort of compelling content that makes people talk about it and spread the word.

This gets away from the class-based idea that there are “basic users” as opposed to “premium users” and then everyone else is a “freeloader.” I felt that this mindset poisoned the atmosphere at ChinesePod whenever the issue of user class and access to content came up.

We operate on a much simpler subscription model.More below...

In short, the point isn’t to be free of advertising. To the contrary, the point is to use the advertising model in place of the paid subscription model in order to give all users free access to all content.

I understand. The fundamental question is about costs. If we could operate the site entirely with UGC and be confident we were still producing the best learning content out there, our costs would fall significantly and it might be feasible to move to an advertising-only model. As many other discussion threads here reveal - people want high quality learning materials. Producing these requires an investment in professional sound recording equipment, etc. I do not believe UGC will be a core feature of the site for most users.

Teacher-generated content will be created though. The real advantage of the Popup system is that it makes it absolutely trivial to create content. I'd even go so far as to guess that the force multiplier on time saved is somewhere in the 10-20x region. Put another way, someone creating a lesson will take 10-20x longer to generate the equivalent materials themselves (pdf, mp3 files tagged with transcripts, photos, etc., tests) that they will using our system. This is especially the case if support for customized traditional, simplified and pinyin transcripts are demanded - as they should be. Using the site to produce learning content also gives people things like manually annotated popup texts that are very difficult to create otherwise. Teachers will not produce these materials unless they are easy to produce.

The importance of an EASY process for lesson creation is really critical. And not only for content creation. If people want to edit their lessons, with Popup Chinese, all they need to do is make the change and hit a single button ("refresh") to regenerate all of the content and reupload it to our servers for distribution.

I think the emphasis on UCG is a distraction here, since it implies that content creation will happen by non-professionals. We see lowering the costs of lesson creation and publishing as something that tilts the market away from mass podcasts and more towards custom lessons generated BY teachers FOR students.

This is why we're so focused on usability, usability, usability.

[trevelyan]

We've just put up a video showing off some the features of the site from a user perspective. If anyone hasn't had the chance to sign up for our beta premium account (go to http://popupchinese.com/voucher and use voucher number "2008AOYUN") you can check it out here:

http://www.youtube.com/watch?v=kF4I8axmsZw

The video focuses on demonstrating the kind of customization built into the system. These features work for every lesson, regardless of who has created it.

As a bonus, you can also see the absolutely hideous placeholder logo we were using before Monday. The font came from dafont.com (fantastic site though!) and reduced our designer to tears.

[leosmith]

A quick note that we've just published our first intermediate podcast for Popup Chinese. Feedback is very welcome. If not already signed up, be sure to use voucher number 2008AOYUN on page http://popupchinese.com/voucher if you'd like free access to the premium materials.

Overall, it was a great lesson. It was refreshing to hear 2 guys do a dialogue for a change. It was also refreshing to hear a podcast come out with a distinct Beijing accent, since many learners are shooting for that. Playing back the dialogue sentence for the vocab example was really nice. Explaining the grammar points really well, in english, made it very clear for me. And having 3 native speakers (2 men for the dialogue, and a woman for the explanations) added a level of legitimacy that a lot of the "high dollar" podcasts don't even have. And as I've said elsewhere, just because of the fact that you are the only modular intermediate podcast that uses more than a word or two of english, you are bound to get a lot of hits. This format is almost exactly the same as the chinesepod intermediate podcasts, which has a proven success rate.

Some things that would improve the podcast for me.

1) A quick line for line translation following the first play of the dialogue. Dialogues are tough for intermediate learners, and a line for line makes them much more comprehensible in my experience. Unless your goal is to force people to read transcripts. In that case, ignore what I just said.

2) Not so much low usage vocab. I felt like some of the words went beyond intermediate.

3) Less english in the vocab explanations. The english speaker really dominated this part, whereas in chinesepod Jenny is definitely the mistress.

4) In the vocab explanations, instead of repeating the dialogue sentence twice, have the dialogue sentence plus a new sentence or two.

5) Add some silly banter. It really makes the podcast more fun.

6) It feels like you repeated the dialogue one too many times. Maybe it's because I've been listening to too many podcasts lately, but I'd prefer 2 times (3 if you count the line for line).

7) Make the free content downloadable without having to join. I'll actually use it if you do. None of the 3 excellent sites that I've been posting about require membership for downloading podcasts.

Edited September 15, 2008 at 07:24 AM by leosmith

[Luobot]

Feedback is very welcome

Some noticeable problems with the audio production

5) Add some silly banter. It really makes the podcast more fun.

... but makes the podcast tedious to listen to more than once.

7) Make the free content downloadable without having to join. I'll actually use it if you do.

That’s why I was advancing the advertising model. Otherwise, can you think of a good incentive for a commercial organization to want you to use their expensively produced content for free?

[leosmith]

Some noticeable problems with the audio production

Agree, but I assume the high quality mp3 that they mention for members will be high quality. I didn't test that because I was too lazy to join. Or are you talking about the member's version?

5) Add some silly banter. It really makes the podcast more fun.

... but makes the podcast tedious to listen to more than once.

Disagree.

7) Make the free content downloadable without having to join. I'll actually use it if you do.

That’s why I was advancing the advertising model. Otherwise, can you think of a good incentive for a commercial organization to want you to use their expensively produced content for free?

Yes. You might ask yourself why the others allow it.

[Luobot]

Some noticeable problems with the audio production

Agree, but I assume the high quality mp3 that they mention for members will be high quality. I didn't test that because I was too lazy to join. Or are you talking about the member's version?

Yes, I was speaking about the audio production quality of the members version that I downloaded.

You might ask yourself why the others allow [free content downloadable without having to join].

1 - Because ChinesePod started it, and others needed to compete.

2 - Because this replaces the need to have an advertising budget when you're starting out.

3 - Because there's a chicken and egg factor to building up a user base, and free content serves as bait to get it started.

4 - Because once they build up their subscription base to the point of diminishing returns, then it's served its purpose, and they drop it (as ChinesePod did).

However, I just read the following comment by Trevelyan on the Popup site, so perhaps there's another twist to their business model that he can explain a bit more.

Our business model is developing an online/offline hybrid that charges for the parts of the service that require customization and keeps the mass products including the podcasts themselves free.

So the PopupChinese business model may be evolving into something more unique and interesting than the repeat performance of ChinesePod, which is what I had initially (perhaps falsely) assumed.

The most basic question is, what products/services are considered custom versus mass other than the podcast?