Chinesepod draws the line

[Luobot]

ChinesePod’s Newbie level is free so there’s no point doing that, as you said. From Elementary on up, however, isn’t free. If this starts at too high of a level then there will be relatively few people who will benefit from it. Plus it makes more sense to start low and gradually scale up. The real sweet spot in the market is Upper Elementary, i.e., the gap between Elementary and Intermediate. ChinesePod is missing that level altogether, so I suggest focusing there initially, and then gradually growing it.

The format I like the most is something like the following:

1 – Briefly say hello, who you are, and what you’re going to do.

2 – Introduce the key vocabulary.

3 – Have a dialogue.

4 – Discuss the dialogue, expand on the vocabulary and word usage, delve into the sentence structure and patterns, and bring up interesting cultural aspects.

5 – Review what’s been covered.

6 – Say goodbye.

Note that item 2 can be combined with item 4, but I find that previewing the vocabulary makes listening to the dialogue more efficient.

My oral skills – or the lack thereof – could only serve as an example of what not to sound like, but I’d be happy to contribute in other ways. For example, I can help out with editing the audio files, possibly proof-reading transcripts, and I can certainly contribute suggestions.

[Luobot]

ChinesePod, to their credit, created a huge space in the free podcast Mandarin language learning market. And one should not begrudge them the revenue they need to continue producing the great podcasts that have made them famous. But by largely (though not entirely) withdrawing from the space that they created, they have now left a vacuum. Nevertheless, ChinesePod has said that they are not concerned that they’re giving an opening to their competition. Why? Because the competition, they assume, can’t succeed at a model that they themselves have apparently failed with and given up on. And, indeed, one might legitimately ask: “So what if this gives the competition an opening … the opening to do what?” (Hmm, sounds like a question I read somewhere …)

While we’re on that question, let’s also question why companies and individuals create and give away their software. And, further, why do they open source it? It makes no sense from the traditional commercial way of looking at things. Yet it just so happens that some of the best software on the market is free, and the free model is still going strong.

Let’s also remind ourselves that Microsoft, at one time, couldn’t conceive of any model other than their own, the one that they used to beat IBM with and then used to roll over their other competition. Then along came Google with a whole new approach. Suddenly free makes sense. But didn’t ChinesePod try this very free model for three years before giving up on it? Well, yes and no. The podcasts were free, while other lesser valued services were relatively expensive. So let’s play devil’s advocate and ask the question: Was ChinesePod’s approach the only way of doing a free (not so free) model? And does everyone, like ChinesePod, need to have 40 extraordinarily talented staff on the payroll to produce a half-way decent podcast with a transcript?

Perhaps someone will follow in Google’s footsteps and create a high quality, free Mandarin podcast site that earns its revenue from advertising. When a site attracts many thousands of eyeballs, it starts to look pretty juicy to advertisers. A site like Chinese-Forums, for example, could leverage its existing user base and attract many thousands more members by being a quality, open podcast site incorporated into the Forums, itself. The two are, in fact, perfectly synergistic. The advertising model is already in place here. Can’t stand advertising at any price? Then you can opt to become a paying or contributing member and join the “no-ads” group.

I started out saying that ChinesePod deserves enormous credit for their accomplishments, but there is now a vacuum. The vacuum could be filled here, and if not, then somewhere else. The point is that others are certain to give it a try, and other approaches may emerge in the process. One thing that I have confidence in is that vacuums are destined to be filled – over and over again, if necessary – until someone finds a way to make it work for both sides of the equation.

[simonlaing]

Well Said Luobot,

We can start it out as a beta and if Roddy shows spike in hits because it perhaps he would share the wealth or we could look at starting another. I do think his unobtrusive advertising model seems to work well, there are other spinoff's possible for revenue as well, but I think the free with advertising model works well for me.

I will check with some of my native chinese speaker friends on their interest and time available to do this kind of stuff.

have fun,

SImoN:)

[renzhe]

Actually, I've often thought that our first episode project can serve as a good substitute/complement for podcasts, at least for the intermediate levels and up.

Instead of a weekly podcast, we have a weekly episode, with vocabulary lists and plot summaries, and they are even graded in terms of difficulty. They are all first episodes, so you can dip in and out, and the summaries can help you out with the hairy spots. And we have 30 episodes already.

Now, if we could lure some of those intermediate and advanced learners who are leaving ChinesePod to this forum, it would be a great asset. Imagine what only 50 new committed users contributing to the discussions could do.

Perhaps this could serve as the "content" to attract people to the forum, like the podcasts you guys are planning? Certainly, if it is presented as "learning content" or some such resource, we'd need to go through the older threads and bring them all up to a certain standard -- many are missing vocabulary lists, and the like. But it already is a great resource, which could serve to bring in new people -- there are obviously many of them around. And, unlike podcasts, this TV-show based learning seems to be rather unique idea on the net.

[Luobot]

This is another good idea. And it's not mutually exclusive with the podcast idea. Both can be done simultaneously.

[jbradfor]

Luobot, you bring up some interesting ideas about open source, but in this case I think there is one major difference between open source software, and trying to do something similar with a podcast.

That difference is that open source works, in large part, by having small contributions from many different people, each improving one aspect of it. In contrast, I have some difficultly imagining how many people can do small improvements to a podcast. This is because once the podcast is done, it very hard for most people to improve an audio file, in contrast to the way that it is fairly each to modify software.

That is not to say it's not possible. One way I could see this working is to have a collaborative development on the transcripts, and once people are "happy enough" (however that is defined), have some native or near-native speakers turn that transcript into a podcast.

And does everyone, like ChinesePod, need to have 40 extraordinarily talented staff on the payroll to produce a half-way decent podcast with a transcript?

Because making a podcast is easy. Making a pedagogically good podcast, OTOH, is not trivial.

And because they tried to re-develop too many learning tools that are already available elsewhere, in an (apparently unsuccessful) effort to get more paying customers.

[realmayo]

they tried to re-develop too many learning tools that are already available elsewhere

Yes, if they introduce a bit of vocab or an expression in a lesson, I don't see the need to be a "premium" subscriber to access CPOD's glossary or additional content -- I can just search sites like www.nciku.com (which I've plugged before, sorry to go on about it...). And if I turn up a blank, I can ask you guys!

& this is a bit ironic, given that CPOD liked to position itself as web-savvy web2.0 cool 'n' funky different from everyone else and, of course, made a lot of the fact that its podcasts were free.

Still, I'm happy to pay 5 dollars a month for access to the podcasts and their transcripts.

[imron]

The thing that Chinesepod has done well is produce a podcast that is interesting to listen to - both content wise, and also due to the personalities of the hosts. Many of the other podcasts out there really don't compare in terms of production value, and often come across as quite plain. With Chinesepod, it's almost really quite similar to listening to a radio show about some topic, rather than a Chinese lesson. Compare this to many of the other Chinese-learning podcasts out there which clearly sound like they're being read from a script in a recording studio.

Actually that's pretty much why I listened to Chinesepod, not to learn Chinese, but rather because it was something interesting to listen to. That being said, I stopped listening regularly when they switched to version 3, and they dropped individual free lesson feeds for the different difficulty levels. As mentioned above, I was only looking for something interesting to listen to, rather than a Chinese lesson, and so the value proposition was never really there for me - especially as I was only interested in the advanced and media podcasts. When they started putting all the lesson types into one free feed, it was easier to find other interesting content rather than delete all the lesson I wasn't interested in. I'd still visit the main site from time to time to see if there was anything that looked interesting, but now that they've closed this section off for non-subscribers I guess I'll stop listening to it altogether. No great loss for them, as I was never a paying customer. No great loss for me as there is plenty of other interesting stuff to listen to.

[Luobot]

Jbradfor, thanks for your suggestions. In discussing free models, my main purpose in mentioning open source was to point out the fact that the free model, as a model in general, can and does work. It works despite the fact that contributors give away their intellectual property and labor for free. It doesn’t mean that the masses have to contribute to the refinement of a podcast. It just means that there is a core group of people who take responsibility for making it happen in order to benefit a community. Regarding how a collaborative work effort could work, let’s take your suggestion as a starting point: “One way I could see this working is to have a collaborative development on the transcripts, and … have some native or near-native speakers turn that transcript into a podcast.” This is very doable. And other volunteers could contribute to other specific tasks that need to be done. For example, audio editing is a breeze for me, so I volunteered to do that. There may be some html coding and other software development that others may wish to contribute over time. Then the open source model starts to look very familiar. I’m certain that a free model is doable and sustainable, and if ChinesePod isn’t going to do it, then someone else surely will. Doing it here at Chinese-Forums has the advantage of leveraging a good sized user base, with many long-time contributing members, and an existing advertising model that pays the bills.

Cost is obviously a big part of the equation in ChinesePod’s thirst for revenue, so I’ll reiterate that regarding their “need to have 40 extraordinarily talented staff on the payroll to produce a half-way decent podcast with a transcript,” is something I just don’t buy. My point here is that ChinesePod’s overhead is, potentially, a serious competitive disadvantage for them, but it certainly doesn’t mean that their costs in any way reflect what it takes to produce a good podcast.

Having said that, however, I do want to agree with Imron’s point that ChinesePod’s dialogues are interesting and well produced. But, then, Imron isn’t going to pay for them, and he doesn’t even listen to them, anyway, so let’s continue talking about the free, community-based model.

All ideas are welcomed.

[realmayo]

All ideas are welcomed.

Any podcast done here would have to be at least slighly better than the several, very good, alternatives out there, otherwise I wouldn't bother listen.

Also, if you're positioning it as a chinese-forums thing, I imagine you'd want Roddy's blessing.

However -- a podcast discussing some of the more interesting threads of the past week or two on these forums is something I'd probably listen to if it was done reasonably well.

[Luobot]

a podcast discussing some of the more interesting threads of the past week

That's a very interesting idea. Keep them coming!

[jbradfor]

If you are seriously considering this, two more items I think you need to address first.

First, the level. [i know this was mentioned, but not decided.] At whatever level you aim this, people are going to be unhappy, either too hard or too easy. But someone did mention that Upper Elementary was an area CP is lacking. [something that has been mentioned quite often too on CP forums.]

Second, once the level is decided, the vocabulary. This is, in my opinion, by far the #1 problem with podcasts as a way of learning. If you know every word in the podcast, you don't learn new words. [You might learn some information, or improve your listening skills, but no new vocab.] If you know too few words, you can't following the discussion, and again you don't learn much. So what I would recommend is creating a list of "core words", words that everyone listening is expected to know. Publish this list, so we can all learn the ones we don't know. Try to keep almost all the words used in the podcast from this list. Then create a second list, "words to learn". This is a list of words that you don't expect everyone to know all of them. Again, publish this list so we can learn them, but try to keep the usage of these words to, oh, I don't know the correct number, but let's say one per every sentence or two. That way we can slowly learn the new works. Lastly, for words not on either list, try to limit them to only a couple per podcast. [Note that if you want to aim at multiple levels, this gives a very clean way of doing so. The "core words" for a given level is the "core words" + the "words to learn" for the previous level.]

Last idea, regarding the format. CP is aimed at learning Chinese. C-F.C is aimed more at information about China in general. This gives you a difference right there. So, for example, rather than focus so much on vocab and grammar, in which the content itself is more a vehicle, make the content itself part of the reason for listening. As imron said, "that's pretty much why I listened to Chinesepod, not to learn Chinese, but rather because it was something interesting to listen to." That should be the goal.

a podcast discussing some of the more interesting threads of the past week or two on these forums is something I'd probably listen to if it was done reasonably well.

Agreed. You have a built-in list of topics, such as "why 17 year olds coming to China to get drunk are totally obnoxious".

[mikeedward]

What annoys me about CPod:

Horrible website design, especially for new users. No quality control. Sheer incompetance in testing. Likes to hand out projects to staff that are a complete waste of time. Half-brained ideas are never fully realized. They say they listen to the users but only hear what they want to hear. Full-length transcripts, traditional character support, slower gradations in level, have all been asked for for years. Full-length transcripts could have closed the lesson gap years ago but the management is blind and/or stupid. CSLPod and iMandarinPod are doing it all for free and have no financial motivations.

400-500 podcasts at newbie level containg 95% English is just plain stupid - are they really trying to teach us Chinese or just earn money? Random podcast topics are for THEIR benefit, not ours. It takes much less effort to make a new random podcast then plan a linear course like chineselearnonline.com. Having never heard Ken (or any other partner) speak anything beyond very basic Mandarin with poor tones, should I really take his word for it that random lessons with no reinforcement of vocab and gratuitous amounts of English is a good way to learn Chinese?

But they hire GREAT people. Pasden, Clay, Jenny, Aric, Amber, Colleen, David L., David Xu, Connie, and the folks behind the scenes. Why do they appear to be failing? They overspend on staff, office space, weekly public-relations gaffes, failing to improve on what people have been asking for and implementing completely the wrong features, horrible website design which probably has a decade of man years invested in it yet for the good of the company should be scrapped. They are fascinated and blinded by new technology and buzzwords - web 2.0, iphones, fancy website crap. It all comes down to horrible management. What do I think about the move to paid podcasts? Should have happened two years ago but you wasted time building a horrible website, expanding, and generally ignoring the wishes of the majority of your users.

Beginners who want the best low-cost path to learning Mandarin cheaply on the Web and without the technical headaches and gratuitous banter of CPod:

Start with this site: http://www.ctcfl.ox.ac.uk/Chinese/lessons.htm

Do the 90 Pimsleur lessons if you can find it/buy it.

Go through the chineselearnonline.com lessons. Do Serge Melnyk for variety.

Do CSLPod.com. Awesome stuff. Do some iMandarinPod.com on the side. Master character reading while you're at it with their full transcripts.

David and Helen http://classes.yale.edu/chns130/Listening/index.html

Orlando Kelm's interviews: http://www.laits.utexas.edu/orkelm/chinese/index.html

Through this path you'll be exposed to much less English, much less distracting conversations about whether people like the lesson of the day or not and the latest CPod public relations disaster, and much more variety in voices (both accent and sex).

Rant off

Edited September 6, 2008 at 03:45 AM by mikeedward
website typo, name typo

[leosmith]

I’ve actually been thinking about creating a podcast site for a while now, for a certain language with few resources. So let me share with you some of my opinions about podcasts.

First, I prefer to use podcasts at the intermediate level. This is because there are many very effective beginner programs that I feel work better for me than a podcast. At the advanced level, I want to concentrate on native material. So if it were me, I’d focus on intermediate. (As an aside, most of what passes for advanced in podcasts I suspect is more like upper intermediate. If we’re not understanding 95% of the vocabulary in a TV broadcast, for example, we probably shouldn’t be calling ourselves advanced. But I think having that extra level makes people feel good about themselves, which is good.)

Second (this is in line with what Jbradfor touched on.), let me illustrate something by dividing podcast sites into four types (please forgive the over-generalizations, and the exclusion of grammar in my descriptions).

Type A: Teaches vocabulary (maybe 1000 words) through a series of progressive podcasts. By progressive, I mean each assumes you have done the preceding lesson. Each lesson translates/explains only the words that haven’t covered before.

Type B: Beginner Level teaches the beginner vocabulary (maybe 1000 words) through a series of modular podcasts. By modular, I mean unrelated. Each lesson translates/explains all the words in the podcast. Lower Intermediate level assumes that all the beginner vocabulary was learned, and teaches lower intermediate vocabulary (maybe an additional 1000 words) using modular podcasts. Each lesson translates/explains all the lower intermediate vocabulary. Intermediate, Upper Intermediate and Advanced will follow the same pattern, each assuming all the previous level’s vocabulary has been learned, and translating/explaining all words in the current level.

Type C: Beginner Level teaches the beginner vocabulary (maybe 1000 words) through a series of modular podcasts. Each lesson translates/explains most of the words in the podcast. Lower Intermediate level assumes that all the beginner vocabulary was learned, and teaches lower intermediate vocabulary (maybe an additional 1000 words) using modular podcasts. Each lesson translates/explains some of the lower intermediate vocabulary. Intermediate, Upper Intermediate and Advanced will follow the same pattern, each assuming all the previous level’s vocabulary has been learned, and translating/explaining some of the new words.

Type D: Essays in the target language. It is assumed the learner already knows basic vocabulary (maybe 1000-3000 words).

(If you were wondering, in these definitions, I consider everything in a podcast to be part of the podcast, not just the dialogue. New words introduced in the banter are still new words.)

Chinesepod does so many things well. They have received a lot of praise here, and they deserve it. But let me be the bad guy for a minute, and focus on things I don’t like about them. Chinesepod does intermediate poorly IMO. I’ve said it many times, but never elaborated. Chinesepod is type C. In fact, they are what I would call a “hard” C, with lots of unexplained intermediate and advanced vocabulary in their intermediate podcasts. Let me list some possible reasons.

1) They don’t have to do the work required to make sure they’ve translated all the unknown vocabulary, and they don’t have to be careful about how they explain things

2) A higher percentage of learners will feel pressure to buy transcripts to get translations

3) A higher percentage of learners will feel pressure to actually read and look up words, improving their learning

If it were up to me, I’d do type B.

(more tomorrow)

[Luobot]

The following summarizes ideas gleaned from this thread. It’s a work in progress, which I’ll edit as new suggestions are brought up.

Podcast Principles and General Guidelines:

- The main focus should be Upper Elementary to Upper Intermediate.

- The podcast should emphasize being “interesting to listen to” over being educationally rigorous.

- Podcast topics should be chosen from the Chinese-Forums threads.

- Use a “modular” approach. (For the definition of “modular,” see post #39 and also posts #34 and #37 for a full description of podcast types.)

- Transcripts should break vocabulary down into separate sections that include: (1) Words you should know, and (2) Words you need to learn.

- Keep the new words per sentence to a low ratio.

- Transcripts should support both simplified and traditional characters.

- The speakers should be primarily native and near native, but may also include guest appearances by C-F members.

Edited September 7, 2008 at 01:32 PM by Luobot

[woliveri]

I'm in agreement with MikeEdward's post with the exception of the Pimsleur recommendation. I think that course is just above worthless.

[leosmith]

If it were up to me, I’d do type B.

(more tomorrow)

And by that I meant I would prefer listening to type B, and would therefore strive to create a type B podcast. And that's even knowing it's the hardest type to create, from a pure accounting-for-vocabulary standpoint.

Type A, while very efficient, feels very restrictive to me. It's nice to have the option of jumping around. If there are 50 intermediate type B podcasts, you could do any of them on a given day. With type A, it's always determined which one you need to do. If you skip one, you'll pay. Most of us have had that kind of pressure before, and we're used to it. But if it can be avoided with any ill effects, why not do it? A modular podcast gives you this option.

Type C, which might be a good option for an advanced learner needing to bring her listening skills up to the level of her other skills, is brutal for a learner who wants to learn by listening alone. It almost forces the learner to buckle-down and read transcripts, possible paying instead of taking the free ride. If there is new material, I want it explained in the podcast.

Type D is fantastic for pure, care-free listening. Studies have shown that if you know 95% or more of the vocabulary of something, you will be able to learn new words most efficiently merely by consuming. The hard part is finding essays in which you know 95% of the material. And if you know too much, you aren't learning much new stuff. For this reason, I consider type D better for "language usage" than language learning.

As I mentioned earlier, all these examples are extreme. Type A podcasts will often cover the same material in more than one lesson. Type B podcasts will sometimes throw in advanced vocabulary without explaining. Type C podcasts will sometimes explain every single word. Type D podcasts will explain material sometimes. etc, etc, etc.

(Tonight I will explain why japanesepod101 is so much better than any other podcast I've tried.)

[imron]

Orlando Kelm's interviews: http://www.laits.utexas.edu/orkelm/chinese/index.html

I'd forgotten how interesting these were. Thanks for reminding.

[Luobot]

Whether the proposed Chinese-Forums podcast is linear or modular is a huge foundational decision, so I thought I’d discuss it a bit further and see if there are any reactions.

In general, my own personal preference is the progressive, linear approach, but I’m not sure if it would be the best approach for this proposed podcast for a number of reasons. For one, there are already good podcasts out there (CLO, Melnyks, and others), that provide free lessons that are linear and progressive. Now, if these sites were to go and follow ChinesePod’s example and start charging for their podcasts, then I could see a strong argument for someone to step into that void. But as we sit here, today, that’s not where the void is. More importantly, the progressive, linear approach doesn’t fit in with the idea that the podcast topics should be chosen from the Chinese-Forums threads, which is is the key, unique, differentiator for a Chinese-Forums podcast. Further, many of the Forum members are students who are already following a course, and this podcast is envisioned as a supplement to real courses (and to compliment progressive, linear podcasts), just as the Chinese-Forums, itself, serves as a supplement and compliment to other language and cultural resources.

So I think the preferred direction is as follows:

Modular Podcasts:

The proposed podcasts should be “modular,” which simply means that they will not be dependent upon one other. This will enable the listener to skip around between podcasts just as the member can skip around to different threads within Chinese-Forums, itself. An attempt will be made to explain all the vocabulary used in each podcast (except for vocabulary considered too basic for the level in which that podcast is rated). The vocabulary that isn’t explained in a podcast will still have definitions included in its transcript, whenever possible. Chinese-Forums members will be able to post their questions within the podcast’s thread in order to get additional explanations from other Chinese-Forums members.

Anyone who cares to do so is encouraged to express their opinion.

[renzhe]

The progressive lesson method really requires a well thought-out structure, and can only be any good if done by somebody with a lot of experience in teaching Chinese. You'd essentially be providing a language course, and those are usually taught by professionals.

The modular concept is much easier to implement, IMHO. Still, there should be something to differentiate it from many similar offerings on the net.