From 2010 to 2014 bitcoin was growing by entering new niche markets: drug trade (Silk Road), gambling (Satoshi Dice), trading (MtGox), trading in China (Chinese exchanges).
It have stopped 2 years ago. Next niche markets (internet-trade, remittances, Wall-street) haven't materialized.
That means that benefits of bitcoin over fiat, for user, are not big even in niche markets to justify switching to it from fiat.
Therefore, bitcoin have to be modified, to make better fit for new markets.
How to modify bitcoin in several (sometimes incompatible) directions?
The way to do it is through sidechains.

I've made a thread about it in general section, but was advised that this topic should really belong to Economics section.
So I'm posting the link here. https://bitcointalk.org/index.php?topic=1272599.0

It's market analysis of a technical solution. That's why I think it fits better not to Economics or Technical, but to General section. Maybe I should put a link to this thread in Economics section?

You are right. Making bitcoin better for one case would make it worse for some other cases. That's why we need several forms of Bitcoins, each one is polished for its own use case, with ability to convert these forms into each other, like $100 bills can be converted to dimes and back. Sidechains would let us to do it.

Lightning Network is important. But it is for resolving other problem: scalability, to increase it on many orders of magnitude. It is not needed yet and if the problem I was discussing won't be solved, it will never be needed. If we won't be able to conquer new niche markets, we'll never need more than 8Mb blocks.  As for other solutions, I'm not aware of them. There are some projects aimed at some specific niches, but there is no project that would enable to attack them all. The sidechain is the only one, there is nothing to compare it with.

I believe that transaction volume in USD prices is the best measure to see bitcoin market penetration. People use bitcoins to transfer value. If price of bitcoins halves, they would need twice as much of bitcoins to transfer the same value. That's why number of bitcoin transacted is bad measure of market penetration. The amount of transaction is even worse in this respect, since the same value can be send by one transacton or by 1000 transactions. USD volume is the best measure. (I agree that inflation-adjusted USD would be better, but so slightly better, that we would need a magnifying glass to see the difference between adjusted and non-adjusted graphs. Because the graph shows growth in many orders of magnitude, while $ value have changed on some % for the period). When bitcoin was gaining new markets (i.e. new groups of people were switching from fiat transactions to btc transactions) we saw the USD volume was jumping new step on the staircase. There were six steps so far. I cannot exactly identify for which market which step was responsible, but I can name following markets: drug trade (thanks Silkroad), gambling (thanks SatoshiDice), trading (thanks Gox), trading in China (thanks Chinese exchanges). Each of these markets gave us usd transaction growth in range of order of magnitude. But for the last 2 years nothing of the kind happens. There is some growth, but nothing comparable to the previous staircase. We have stopped gaining new niche markets, and to resume it, we have to change Bitcoin somehow. Sidechains is the only solution of this problem I'm aware of.

The major benefit of fiat is network effect. You can use USD at any place over whole country. Everybody accept them. While to earn or spend, say, Mongolian Tughrik in New York you would have a lot of hassle. The same is true about bitcoin. To justify this hassle, you would need some really good benefit of bitcoin over fiat.

You already have bitcoin. You are old market. While we are talking about entering new markets. So we need to find a new market, i.e. a category of people that don't use Bitcoin now and change bitcoin in such a way that they would want to use it. If this would work, we should repeat it again, for some other and larger markets. Or we can introduce sidechains and do it simultaneously, for many markets at once, in a hope that at least one of them would fire. It would improve chances and speed of the mainstream adoption.
 
You not only want it, you are currently using at least 3 non-competing currencies. I mean 3 forms of dollars: change, bills, electronic accounts. Each form is good for some purposes, but bad for others. Dimes are not convenient to buy a car with, but are good to pay to a street musician. Electronic account is good to automatically pay rent/mortgage, but not good to pay "under the table". And so on. That's why we have those different "avatars" of dollar. Sidechains will create the same system for bitcoin, only better one: more forms of bitcoin, better tailored for each purpose.  

You are right. I assumed everybody here knows it. OK, for the purpose of this discussion it's enough to know what sidechain would let us to develop new forms of bitcoin, interchangeable into each other, like physical 100-dollar bills can be always changed to dimes and back, without losing it's value. There is no limit to number and variety of possible bitcoin forms, so they can be precisely tailored to each market. The switch between them can be handled by your wallet, you may not even need to know about it. You just take from your wallet item you currently need, be it a dime or $1000 bill. When needed, the wallet will break your $1000 bill into dimes or merge the dimes back or convert them to any other form you would need.

There are already technical threads, dedicated to it. This thread is about market analysis. My goal was to demonstrate that to significantly increase bitcoin's market share we need to transform it into an array of non-competing currencies. Sidechains is the only project aimed at it.

Anyway, I'm glad that our positions are closer than I thought.

Yes, money is important. But people already have money system. They need some very strong reasons to discard it and switch to something else. What benefits this "something else" would give to them? Those benefits should be really big, to overcome tremendous inertia and network effect of the current money system. Email did have such a benefit: it delivered mail thousand times faster. What thousand-fold benefit over dollar bitcoin has? For average Joe? In an area that really matters to Joe, like speed of mail delivery mattered?

Updated OP with a multi-colored graph.

Thanks, that's something that we can discuss:


-I see two problem with this argument.
a)  The "there are possibly better solutions" is not an argument at all, because exactly the same can be said about any solutions of any problem in the world. Correct? This argument is faulty because it is impossible to prove non-existence of something. Non-existence of possible better solution, in particular. In a healthy discussion an objection to a suggested solution should be either: "I suggest another solution, it is better and that's why:". Or "Status quo is better than changes you are suggesting and that's why:". If you have such an argument, let's discuss it. It is quite possible that you are right and I am wrong, but you should demonstrate it. That's how healthy discussions go, in my understanding.

b) You put the word 'problems' in quotes. This indicates that you don't believe that they are problems. Why then you are talking about solutions if there are no problems in the first place? You've said I am confused. Now you can see why  So, could you indicate a bit more clearly which my statement you are arguing with? I'll repeat them here, in more detail:

1. Two years ago it was generally expected among bitcoin community that bitcoin will soon get new niche markets: remittance, internet trade, Wall-street etc. Do you agree?
2. This didn't happen. Do you agree?
3. Bitcoin usage is not growing, at least at such extend that it can be distinguished from speculative rallies/crashes. Here is the proof: https://blockchain.info/charts/estimated-transaction-volume-usd?showDataPoints=false×pan=all&show_header=true&daysAverageString=7&scale=1&address= (EDIT: I put the graph with comments in OP)
As you can see, for the period from Jan11 to Jan14, i.e. for 3 years it did grow steadily and had gained 4 orders of magnitude. But for the following 2 years, from jan14 till now, it have not gained not a single order of magnitude. We can clearly see 3-year "staircase" and then 2-year flat line. The only conclusion I can make is that bitcoin was gaining new and new markets, for 3 years, but 2 years ago it have practically stopped. Do you agree?
4. Bitcoin have not (yet) replaced fiat from mainstream market. Do you agree?
5. Same is true about niche market. Even drug trade (the most fitting niche market) is still mostly runs on fiat. Do you agree?
6. Bitcoin is not replacing fiat from mainstream market because it' benefits over fiat it not big enough to justify it for customer. Do you agree? (I hope you do, because it's almost an tautology ).
7. Bitcoin is not replacing fiat from niche markets by the same reason. Do you agree?
8. The way for bitcoin adoption goes through niche market. After conquering a niche market it strength grows and it makes easier to enter the next niche. Do you agree?
9. For each niche market there are some feature(s) of bitcoin that can be changed/added, to make it better fit and increase chances of it being adopted. (Scalability, speed, privacy, taxability etc). You may disagree with specific examples, but do you agree that such niches and features exist?
10. Do you agree that making bitcoin better fit for one niche, we can make it worse fit for some other niches?
11. Do you agree that in situation when we have an array of virtual currencies, each one tailored for its own niche market, with their total value fixed (I.e. that total value of all altcoins is never above 21M BTC and you'll never lose on exchanging from altcoin to bitcoin and back), chances of them getting mainstream adoption would be considerably higher than now?
12. Do you agree that sidechains offers such a solution?
13. Do you know any other solution that gives us the same results: array of niche-tailored non-competing currencies?

I would be glad to make detailed answer to your post, but alas. You've gave me nothing to answer to. No facts, no reasoning, just a fountain of emotions. What one can answer to pure emotions? Calm down, take it easy? OK, please calm down and take it easy.

But if, apart from emotions, you have something else to say, if you have found a mistake in my facts or a logical fault in my reasoning, then I'll be glad to discuss it with you.

It's not my BIP. I'm not suggesting any specific BIP. I'm suggesting a BIP that would enable sidechains. Because I think that without sidechains chances of bitcoin getting mainstream adoption would be much lower. Why I think so - see OP.

Bitcoin is good, but it's not good enough.

Internet have won when emai have appeared. Email was so much better than snail-mail, for everybody, that it have become the killer app and the mainstream adoption have happened.

There is no such killer application for bitcoin. Let's face it, Bitcoin's benefits over fiat are currently not big enough to justify switching to it.
Therefore we have to find a more fitting niche market, with specific needs where bitcoin's benefit over fiat are higher.
There are several such markets.

However, the last year have shown that bitcoin is not taking over remittance market or internet-sales market or PESA market or even black market. Bitcoin is not good enough even for a niche markets.

Therefore, to take over one of these markets, it have to be modified, to make a better fit.
For example:

-For black and gray market it have to be much more private.
-For cup-of-coffee market it have to be much more fast. And much more scalable.
-For PESA market it have to be more SMS-friendly
-For bank backbone market it have to be much more secure (compulsory multisign etc)
-For record-keeping market it have to have more fields for information storage
-For smart-contract market it have to have smart-contract language embedded
etc etc etc

The above are just some rough ideas. Undoubtedly, there are much more possible changes that would make bitcoin much better fit for those markets.
What's more, many of these changes have been already invented and even implemented in some altcoins.

All we have to do is to combine these improvement with benefits of mainstream bitcoin: first mover advantage and security, based on powerful mining factories.

There is no guarantee it will happen. But it depends not on whims of some billionaire-whale, but on us. We have to do just two things:

1) Fight hard to overcome powerful interest groups in bitcoin community and put the sidechains BIP into the bitcoin core.
2) Work hard to implement those niche market sidechains.
      -more-secure-than-bitcoin  sidechain for backbone of banks
      -more-private-than-bitcoin sidechain for black-gray market
      -more-scalable-than-bitcoin sidechain for internet shopping
      -more-fast-than-bitcoin sidechain for cup-of-coffee purchases
      -more-sms-friendly-than-bitcoin sidechain to replace PESA
      -more and more of these specialised sidecoins for niche markets


And then - to the moon.

tl;dr; Bitcoin can win only through a niche market. To win a niche market, we need sidechain BIP.

"Beating itself up" was direct result of war with the krauts. It was a war of attrition: which government would use up their country's resources and collapse. Russia's government collapsed first. Germany collapsed 8 months later, so Russans get a chance to take part of what they gave to Germans back (like Ukraine). The rest (and more) they took back during WWII.

Even Russia have lost one of those (WWI) https://en.wikipedia.org/wiki/Treaty_of_Brest-Litovsk

Bitcoins as such doesn't give much benefits to justify mass adoption.

However, it can gain mass adoption through its sidechains, tailored for different niche markets.
-one sidechain tailored for cup-of-coffee (a lot of non-secure, but easy and fast transactions)
-second sidechain tailored for black & gray market (complete anonimity and untracebility)
-third sidechain tailored for investment (slow but super-secure)
-fourth sidechain tailored for legal applications (wide range of smart contracts)
-fourth sidechain tailored for storing all kind of information
-etc etc etc

If so, we should pray for the success of Blockstream.

For the last two years usage of bitcoin has not changed. Usd transaction volume is stays about the same:
https://blockchain.info/charts/estimated-transaction-volume-usd?showDataPoints=false&show_header=true&daysAverageString=7×pan=all&scale=1&address=
 When a killer usecase will be found, we'll get the growth.

IMO, bitcoin currently has no noticeable benefits for common users. However, there are small, moderate and big benefits for some categories of users:

-You bank won't be able to spy on you or freeze your account. Huge benefit for black and gray market traders.
-You can receive bitcoins just as easy as send them. Moderate benefit for owners of small internet shops.
-You bitcoins aren't damaged by inflation. Moderate benefit for Argentinians, small benefit for 1-st world users after the third halving.

After the second halving, July next year, bitcoin inflation will become about 4%. Which is in about the same as US dollar's. Four years later it will drop below 2%, so it will be some small benefit for most of user. However, this small benefit, it is nothing like email-vs-snailmail difference, which gave boost to Internet. To get mass bitcoin adoption we need some killer app, some killer user case which would give some real and noticeable benefit to most of common users. We don't have it so far.

I wonder myself    Here is the link to the paper. https://scalingbitcoin.org/montreal2015/presentations/Day1/8-Ittay-eyal-testbed-for-bitcoin-scaling.pdf
The description is in the second half, it's called Bitcoin-NG.
I've found some articles about it http://hackingdistributed.com/2015/10/14/bitcoin-ng/, https://bitcoinmagazine.com/articles/bitcoin-ng-or-how-cornell-researchers-think-a-radical-redesign-can-solve-bitcoin-s-scaling-issues-1447108649
And here is the discussion or Reddit https://www.reddit.com/r/Bitcoin/comments/3or365/bitcoinng_proposal_for_a_secure_faster_better/

Why are you so sure? Are you scientist that specializes in this area? If not, how do you know? From scientists? How do you know what they think? Let's imagine for a second that you are a scientist and you think that there is some validity to this idea. Would you announce this thought in public? To become a racist in the eyes of this public? If you plan a research in this area, which evidences would you look for - for and against this idea? What would you prefer - to receive a grant or to become a target for witch-hunt?

So the honest answer to the question "is there any scientific validity to the idea of different human races" should be: We don't know because unbiased research in this area is politically impossible.

2 Mb. Just because it's politically easier to implement. It'll buy us some time.
Time to implement something different, like lightning.

My favorite one - the solution that was suggested at the conference: introduce the second type of blocks that contain transactions only and issued 60 times more often, every 10 seconds or so. Those blocks are not for determining winner of reward, so they won't cause orphan problems. This solution will give us 60 times more space for transactions and reduce confirmation time 60-fold!  I think it is the perfect solution, even better than lightning, since it's not hub-based.

Sometimes he does, sometimes he doesn't. For example, two not-quoted graphs on this page:

Then, as a suggestion, you can put references to origin of each TA post, so people can see discuss / read discussion about them.