Glossika method

[OneEye]

I don't know the details of Mike's business, but having looked into ebook publishing a bit, I imagine Amazon takes a generous cut, and perhaps whatever ebook publishers and distributors he works with get a cut too. Odds are, Glossika really doesn't get much per sale. It's not just about bandwidth, there are a lot of other factors. So though some seem to think they're being shady, I really doubt that's the case.

 

I also don't agree with people who seem to think that they ought to apologize for being Taiwanese by putting a disclaimer on the course or giving a refund/discount to those who want to buy the China version also. It's a Taiwanese company, what do you expect but Taiwanese people and Taiwanese Mandarin? Would you want a British company to apologize for not providing American English in their products?

[realmayo]

Would you want a British company to apologize for not providing American English in their products?

 

Okay but what if they gave you audio with a thick Birmingham accent throughout?

 

I'm not sure that's shady about them. Just not very aware about what most Chinese learners want.

[renzhe]

Where do you guys see progress 97%? on what website?

edit: ok i see it on bottom with blue bars.

I still can't find it, could someone post a direct link or something?

About a year ago, I was planning to do something very much like this, for drilling sentences. I was going to do it all by myself, with native speakers reading the sentences I had collected, and by cutting segments of TV shows. I was even playing with the idea of launching a website with a donation box, or some sort of product because it's a lot of work.

Just about the same time, Mike did all of that so I don't have to. As soon as the Mainland version is finished, I'm buying it.

If they finish a European Portuguese version, I'm buying that too.

[realmayo]

http://www.glossika.com/

 

scroll to the bottom and more page will load, keep scrolling, eventually you should get it

[OneEye]

Okay but what if they gave you audio with a thick Birmingham accent throughout?

 

The recordings present a very clear, neutral, normal Taiwanese accent, i.e., the way most young, educated, native Mandarin speakers in Taiwan speak. So I don't really see what you're getting at here.

 

I personally don't trust "what most Chinese learners want," because in my experience what they want is something to make them feel good about their efforts, not necessarily something that's effective, especially not if it's difficult or uncomfortable. And I don't think Mike's goal here was to take the Chinese learning market by storm, but the overall language learning market. He's really aiming at the polyglot crowd here. He's providing authentic material with authentic (not "trained," but authentic) recordings for people who want to learn a language "as she is spoke" rather than "as she is taught to rooms of foreigners who don't yet know any better."

 

I'm guessing that's why he works with translators rather than language teachers, because in my experience translators have a much higher standard for authenticity of language use. Teachers think stilted, bookish Chinese is wonderful because it's "advanced," but translators wouldn't put up with it if it's unnatural. I think it's admirable of Mike to insist on this kind of quality, and it's something sorely missing from the glut of over-produced, inauthentic, stilted language learning textbooks out there. And in his quest for real language, being based in Taiwan and being an exceptional speaker of Taiwanese Mandarin himself, he sought out a Taiwanese person with good Mandarin to collaborate with on his Mandarin course. If anyone was expecting CCTV putonghua, they didn't put much effort into finding out what they were buying.

[realmayo]

First: I don't want to be especially critical: I have bought some of the Glossika things, including one of the Mandarin ones, I think they are excellent resources and would recommend them: as renzhe suggests, the effort of creating something similar on one's own would be immense versus the expense of buying these products pre-made.

 

 

But why's it so hard to see that most learners will want the mainland standard, not the Taiwanese one? On the mainland, foreigners and Chinese people alike know that there is the somewhat artificial CCTV-ish standard spoken on TV and in textbooks, and there is the Chinese that is used in real life. Depending on where you live, what you speak will differ by a greater or lesser extent from the Mainland standard.

 

But any concentrated use of the Glossika materials means learning to speak in a third accent. That's tiresome.

 

Definitely, exposure to different accents is important and fun and good. That's one reason I like the 锵锵三人行 TV programme for learning. But there's a difference between exposure and intensive work.

 

 

As for whether these materials are authentic or not, that's not really relevant here. This isn't a question about the content of these sentences, it's about accent. And of course it's not "authentic" to hear the same voice in your head for 3000 sentences at a time, so I don't see what authenticity has to do with it at all.

 

 

Note that the course says: 

Source Language: English (American)
Target Language: Chinese (Mandarin)

 

So they distinguish between American English but not Taiwan Mandarin.

 

 

Also I think comparing British/American English with Mainland/Taiwan Mandarin is wrong. Taiwan is a small country with minimal influence and its version of Mandarin has minimal influence outside of Taiwan. But for English, if you live in (say) Kazakhstan, you'll typically choose between the two main standards, US and British. British English remains a standard not because of the current importance of the UK in the world, but because historically that's the standard that a lot of European learners have followed. 

 

But you wouldn't choose to follow the New Zealand accent (maybe it's better if I say New Zealand not Birmingham). Unless you lived in New Zealand. In which case, hopefully you'd end up being a fluent speaker with a slight NZ accent which would cause you no problems wherever you went.

[gato]

One must admit that the recording at the moment is really 嗲.

[etm001]

@imron said:

 

 

A terabyte of bandwidth should cost anywhere from $50 to $200 a month. So 2-3 people purchasing the product would cover 1,000 1gb downloads.

I don't know anything about bandwidth costs, but I can say this: the Glossika files are encoded as 128kps/44.1kHz joint stereo MP3s. This seems excessive for spoken audio, i.e., using a lower bit/sample rate and a mono channel would produce smaller files without sacrificing audio quality. (See also posts #201 and #202 in this thread, in which it was noted that the use of cue sheets might further reduce file sizes).

 

@Snuggles said:

 

 

 I have access to plenty of TV from Taiwan. Also, I live in China. I hear quite a range of accents each day. Plenty of people hear speak quite differently from the standard.

 

@JustinJJ said:

 

 

I agree. For listening it's easy to get exposure to various accents. For imitation of an accent, just learning the standard you want to speak should be sufficient (unless you specifically want to be able to speak in various accents).

 

Again and again in this thread people keep coming back to "accent" (using it as a general term ecompassing both accent and dialect)

[realmayo]

BAgain and again in this thread people keep coming back to "accent" (using it as a general term ecompassing both accent and dialect)

 

Well, is it that the Mandarin regarded as the official standard in Taiwan is a different dialect to that deployed as the official standard in Beijing?

[imron]

I imagine Amazon takes a generous cut

That would only be the case if they were selling ebooks through Amazon, which they're not (at least not from their homepage).  The payment platform is Paypal (which will take maybe 4-5%).  Files appear to be hosted on Amazon's web services (totally different costs compared to them being the publisher/distributor) which is where the bandwidth costs I was talking about earlier come in and although Amazon is pricier compared to some other providers, it should still be within the range I mentioned above.

 

I agree however that there are costs elsewhere - partnering agreements, printing copies of the books (not sure if there is huge demand for this over ebook versions), development of materials, affiliate program etc.  My previous post was not saying that the products were cheap to produce, it was just countering a previous point made by someone else saying that bandwidth would be a major cost.

[hedwards]

@imron, that assumes that they have a large profit margin. This is a niche product, sort of like your own, I'd be very surprised if they were selling tens of thousands of units. I get the distinct feeling that they aren't making that much money per package as it's rare for any product I buy to have such a limited period of time during which to download it. In fact, I don't think I've ever bought one that didn't give me at least a month of unlimited downloads. Even a week is rather unusual, most of the time you get to download forever, or at least for the length of the term of service.

 

 

Anyways, I really like how people seem to think that that $70 is pure profit. Eventually, as the cost of producing the materials, hosting their site and correcting their materials is reduced, it comprise more and more profit, but there's other expenses that they have to fund with that money. And don't forget that we do get updates to the program for free, so they're not going to be recouping those costs from people who have already bought the product.

 

Ultimately, it's self-indulgent and self-entitled to expect that because there's a new program out that's closer to what you want that we should get a discount on that. They already gave us a discount on a discount, and now people are complaining about not getting another discount on top of a discount like the last time.

 

Anyways, I've got better things to do than to continue this. Without access to their books it's only going to be speculation, but it's ridiculous to assume that because the product is digital that they've already been compensated for the sunk costs associated with creating that first copy.

 

EDIT: According to their page they have 10,000+ customers, that's spread across a relatively large number of products. No wonder they have a tiny profit margin at this point.

[imron]

Bandwidth on AWS is metered, so you pay for what you use.  If you only have 1GB of transfer you'll only pay for 1GB.  My point was not that they were doing large volumes, just that a single purchase covers the cost of a large amount of downloads.

 

The limited time for download is also a feature of AWS - auto-expiring links, and is not for bandwidth limiting, but a form of content protection.  You'll note that you never had to enter a password or anything to download the material, that's because the links aren't in anyway protected.  Instead they'll be unique (and unguessable) for each user, and set to auto-expire after a fixed period of time meaning you can't just pass on the link to someone else.

 

Is it worth doing something like that?  Well, it's debatable, but Glossika seems a bit paranoid in this regard - see also PDFs not containing text, which provides a degraded user experience (no searching, larger downloads) in what is presumably an effort to prevent people from copying and pasting content elsewhere.

 

Compare it to say ChinesePod, which a few years back (and possibly still now) had a bug where they were using guessable links so you could download all the material you liked if you knew a subscriber's username (and all the hosts had premium access and publicised names).  I told them about it at the time, but they didn't seem to care.

 

For your other points, we are not in disagreement.

[Auberon]

 

Well, is it that the Mandarin regarded as the official standard in Taiwan is a different dialect to that deployed as the official standard in Beijing?

 

This is apparently a complete list of different pronunciations in official standard ROC Guoyu compared with PRC Putonghua:

 

http://www.zhongwen.com/x/guopu.htm

 

There aren't many, and the vast majority consist merely of a difference in tone. Of course, this is only the official standard, not the colloquial reality, but the colloquial language varies enormously throughout the mainland, too. The ROC standard itself came from the language of officials, and before they were forced to relocate to Taiwan, they would need to have spoken Mandarin that would easily be understood by Beijing; that is the point of having a standard.

'Dialect' is a nebulous term in itself. There is no accepted definition of what constitutes a dialect with regard to comprehensibility by speakers of an 'official' version of the language. 'Accent', however, merely means pronouncing the same word slightly differently, which fits the difference between the official standards of the ROC and the PRC.

 

 

But why's it so hard to see that most learners will want the mainland standard, not the Taiwanese one? [...]

But any concentrated use of the Glossika materials means learning to speak in a third accent. That's tiresome.

 

But almost every course uses the mainland standard. There are a thousand and one of them for anyone to choose. I think it's refreshing to see resources for Taiwanese Mandarin, as there aren't many of them about, and Taiwan is by no means globally insignificant.

[Snuggles]

But it should be noted in the course that it is the Taiwanese version. People purchasing English materials would be aware of both British and American materials and purchase what they want. However for people who purchased this product early on, they would have been unaware (and rightfully so) that it was not the standard mainland Mandarin.

I have bought lots of Mandarin materials. Not a single one was Taiwanese. It would be proper to have my copy replaced with the correct version when it comes out. I don't think that is entitled thinking, but rather running a responsible business.

[Auberon]

But it's still Mandarin. It isn't a different language.

 

 

But it should be noted in the course that it is the Taiwanese version. People purchasing English materials would be aware of both British and American materials and purchase what they want.

 

But if an English course were described simply as 'English' (and they frequently are), then you don't know what you're getting. If you make an assumption, it is hardly the fault of the maker. It's still English; you will still be able to use it to communicate whichever part of the English-speaking world you happen to be in. If you only want British or American or some other variety of English, then only buy materials that state it.

For another example; I like my Chinese learning materials to be in traditional characters. So I don't, as a rule, buy books that aren't explicitly described as including them. When I break that rule, and end up with something in the other character set, I don't consider myself entitled to a refund or discount.

[imron]

However for people who purchased this product early on, they would have been unaware (and rightfully so) that it was not the standard mainland Mandarin

I'm not sure that was the case.  It's definitely been discussed in this thread since the first page so anyone reading this thread before making a purchase decision should have been aware of it.

 

There were also examples of the content available on the website prior to purchase.

[JustinJJ]

But not everyone buying the product would read chinese forums, so it would have been easier to assume they were buying a mainland standard

 

I don't think it's fair to compare English with Chinese because in English there is no one standard (at least I've never heard of one). However in Chinese there is definately a standard and *all else equal* the majority of students of Chinese language are going to assume 'mandarin Chinese' is the mainland standard.

 

Without knowing anything anything about their business model, my two cents is that if the company were to offer discounts or free mainland versions to early purchasers it could be a way to increase their brand image and entice early purchasers to buy additional products or recommend it to other students. If the mainland version is more useful to an earlier purchaser of the taiwan version, they'll use the mainland version more and recommend it. If they started with the taiwan version and found the taiwan version of '和' griped (for example), they would probably stop using it and not recommend it to others.

 

I suspect that if the business is running efficiently (e.g. not paying over market rates for bandwidth, etc) the marginal cost to selling an additional product shouldn't be high after the product has been produced already. I would have thought the largest costs would be paying people to record the material, equipment, time spent writing the material, editing, etc.

[etm001]

Tl;dr: people like to make mountains out of molehills.

 

@imron said:

 

 

A terabyte of bandwidth should cost anywhere from $50 to $200 a month. So 2-3 people purchasing the product would cover 1,000 1gb downloads.

I don't know anything about bandwidth costs, but I can say this: the Glossika files are encoded as 128kps/44.1kHz joint stereo MP3s. This seems excessive for spoken audio, i.e., using a lower bit/sample rate and a mono channel would produce smaller files without sacrificing audio quality. (See also posts #201 and #202 in this thread, in which it was noted that the use of cue sheets might further reduce file sizes).

 

@Snuggles said:

 

 I have access to plenty of TV from Taiwan. Also, I live in China. I hear quite a range of accents each day. Plenty of people hear speak quite differently from the standard.

 

@JustinJJ said:

 

I agree. For listening it's easy to get exposure to various accents. For imitation of an accent, just learning the standard you want to speak should be sufficient (unless you specifically want to be able to speak in various accents).

 

@realmayo said:

 

But any concentrated use of the Glossika materials means learning to speak in a third accent.

 

More than anything "accent" seems to be the real sticking point for some people in this thread. (To be clear, what people are concerned about includes dialect, accent, and pronunciation). This is surprising to me for a few reasons:

• If you are using Glossika, then you are not a beginning Mandarin speaker (the material is not intended for beginners). You've been learning/speaking Mandarin long enough such that your own accent should be well established. As far as accent goes, you are not going to start sounding Taiwanese by listening to the Glossika materials.
• From the above quotes we see that (at least some) people are exposed to a range of "non-standard" accents in their daily lives and this (presumably) doesn't pose any problems for them. So why is it a problem in the Glossika materials?

It never crossed my mind to use Glossika as a tool to refine my accent. I have no worries that, if I listen to the mainland China-based package that I'll start sounding like a mainlander. But if you are concerned that hearing Taiwanese Mandarin will affect your accent/pronunciation, then a) consider whether Glossika is appropriate for your level of Mandarin, and/or b) wait for the mainland China-based package.

 

What about dialect differences between Taiwanese "standard" Mandarin and mainland "standard" Mandarin? This is a valid question. In one of the Glossika packages you hear the speaker say 我會口渴. Are you likely to hear that in mainland China? (Honestly, I've never heard someone say it in Taiwan either). What about 你打哪兒來? Did you hear that in the current Glossika package? No you didn't, because no one in Taiwan would say that. What you heard was 你從哪裡來 and variations thereof. Was it a good thing or a bad thing to hear those variations? That's up to you decide. Again, if I'm listening to the mainland China-based package and I hear 你打哪兒來, I'm going to be happy. Happy to learn one more (albeit small) way to express myself in Mandarin.

 

Another value in hearing different dialects/accents/pronunciations is that it can help me sound more like a local. It's not something I have to do (and I'm certainly not ashamed of my current accent/pronunciation), it's just something I'd do to fit in a little better. I know some of you are thinking, "99% of the time I'm going to be surrounded by mainlanders, I don't need or want to adjust my words/accent/pronunciation to fit in [Taiwan / another part of China]". And that's fine. 

 

@realmayo said:

 

As for whether these materials are authentic or not, that's not really relevant here. This isn't a question about the content of these sentences, it's about accent.

Authenticity is a main selling point of the product, i.e., hearing authentic (i.e., native) speakers speaking authentic (i.e., natural, colloquial) sentences in Mandarin. To that end, the content of the sentences is of extremely important. It sounds like you are not focused on content but on accent/pronunciation practice? If so then yes, wait for the mainland China-based package.

 

@realmayo said:

 

Also I think comparing British/American English with Mainland/Taiwan Mandarin is wrong.

 

This comparison is made all the time. It's an apt comparison. You're the first person that I've ever heard call it wrong. 

 

@realmayo said:

 

But for English, if you live in (say) Kazakhstan, you'll typically choose between the two main standards, US and British. British English remains a standard not because of the current importance of the UK in the world, but because historically that's the standard that a lot of European learners have followed. 

In most countries to which I've been there is not a choice between English dialects - one dialect is usually firmly established in the area and all the schools teach it. But otherwise I agree with your point.  A person in Kazakhstan can learn British English and not be concerned that they will be unable to understand American English, despite differences such as color/colour and truck/lorry. That same is true learning "standard" Taiwanese Mandarin and understanding "standard" mainland Mandarin. There will be differences, but nothing to which one can't quickly adjust.

 

@gato said:

 

One must admit that the recording at the moment is really 嗲.

It's fairly "standard" Taiwanese Mandarin accent/pronunciation, but yeah I agree it is a bit 嗲. It's not a problem for me but I can see how others might not like it. Whether it's worth all the moaning and gnashing of teeth we've seen in this thread...well, let's just say that when I hear the mainland China-based package, you won't hear me b*tching because it sounds like a pirate talking. 

 

Prediction: someone will complain that the speaker's accent/pronunciation in the mainland China-based package is <insert complaint here>. 

 

@Snuggles said:

 

It would be proper to have my copy replaced with the correct version when it comes out.

What do you mean by "correct version"? Do you mean "correct" Mandarin? By analogy, that's like someone from the UK telling an American that British English is "correct" English.

 

@Auberon said:

 

But it's still Mandarin. It isn't a different language.

Exactly. There's tremendous value in the material even if the speaker is not using the dialect/accent/pronunciation that you prefer. There is not a vast, uncrossable chasm between mainland and Taiwanese Mandarin. And whatever differences there are, I'm sure most of the people reading this thread (i.e., Mandarin and/or language nerds) should have no problem adjusting to them, because they are minor.

 

Finally, some people are upset that they didn't know/realize that the current Mandarin package is based on "standard" Taiwanese Mandarin. I sympathize with that - it sounds like some of you would not have bought it had you know that, and I think it should have been abundantly clear from the start that it was not "standard" mainland Mandarin. (This wasn't a concern for me at the time of purchase; I don't recall how well the sales material did/did not note this).

[MPhillips]

Just an aside about British versus Am. English, there used to exist an accent here halfway in between (Mid-Atlantic) as exemplified by many Hollywood stars (Cary Grant, Ray Milland, Katherine Hepburn etc.) and members of the Am. upper class, but with the death of Gore Vidal it seems to have lost its last well-known speaker (kind of a shame).

[Snuggles]

You mean the English on Glossika isn't the mid Atlantic accent. I shalst be having a word with Mike about that.