You might have a good point about the top 5 traded currencies might be more important than other considerations.. so I would be open to the idea of changing it to the top 5 currencies based on trade volume. Based on that coinhills website, the Canadian Dollar is really far down the list (#18), so it does not seem as relevant based on such trade volume considerations.
You might want to consider Canada's place in the Bitcoin community.
Yes, Singapore may have a greater
exchange trading volume but does that make it more important here in Bitcointalk?
How many of Canada's Bitcoiners use exchanges all that often. Many of us have been involved for so long that we don't trade unnecessarily and when we do there's a good chance it's going to be over the counter or under the table. Most probably simply hodl.
How many BCT members come from Canada as compared to Singapore or some of the other countries on Coinhills' list?
It probably depends on what "we" might be wanting to see in the data - and if they are reflective of actual BTC trading activities going on in the world. I would be more concerned about if trading in any particular currency is fake to be more important to figuring out its non-importance rather than including for reasons that might not be easy to justify...
Do we want to limit the number to 5 currencies?.. is there something meaningful in terms of choosing based on something other than volume? The Coinhill list seems to be trade volume that is
based on all 89 exchanges that are listed through its site; it is not based on any particular exchange.. if there might be some other way of figuring out trade volume but yeah, you are not seeming to like the idea of trade volume in terms of importance...
I won't be taking Canada off this list, but I could see rotating the GBP and JPY with maybe South Africa, and Brazil.
What about the Korean Won? I personally think volume is important, but I am not that attached to the idea, since the main thing that I look at is the dollar... at least most of the time.
~snip~
For some reason the green is missing in this table (the oldest date).
You might have a good point about the top 5 traded currencies might be more important than other considerations.. so I would be open to the idea of changing it to the top 5 currencies based on trade volume. Based on that coinhills website, the Canadian Dollar is really far down the list (#18), so it does not seem as relevant based on such trade volume considerations.
I understand that the top 5 based on trade volume could fluctuate, but on that website it does seem that the top 4 have more of a distance from the 5th place, which likely the 5th place will fluctuate a lot more.... since the 5th one is currently under 1,000 BTC traded.. at least for today...maybe we might consider over a longer period of time in order to analyze if trade volume might change very much in terms of putting which ones into the top 5?
So yeah the top 5based on today's trade volume would be 1) the dollar 2) the Korean Won 3) JPY 4) EUR.... 5) SGD.... I am not exactly attached to any of the non-dollar currencies, but it does seem to be interesting to see what the top traded volume coins might be doing (to the extent that the trade volume is determined to not be fake). Prior to the 2016 China crack down on exchanges at that time, the CNY was trading like 10x anyone else, but then we realized that as with many things chinese, there was a lot of fakeness to it.
I imagine that once we figure out which currencies to retain in the charts, you probably would not want to be switching very frequently.. .. so I am not sure if there might be a better way of choosing which ones to retain, except based on something like trade volume.. and perhaps even better if there were a way to look at trade volume over a longer period of time to see if there might be a consistent set in the top 5... maybe in the last 3-6 months.. or even weekly for the last 3-6 months.
I don't know what you're talking about with the green.
The green is present in the latest table.
|
61 2021-03-13 59670 |
If you look at dooglus' code, that this project was launched with, the green part never shows up. We're talking some real feature creep here.
Translation, I haven't figured that out yet. Back to work.
Even if you haven't figured it out. There is green there, and maybe the green could also have bold.. since it is a little bit difficult to see sometimes, it is not always very obvious that it is green.
Oh and changing the currency is fairly simple now, provided bitcoinity has daily data for the past 5 years.
I am glad you got it figured out... and are you (and maybe everyone else) inclined to think that maxing out at 5 currencies per table is reasonable?
~snip~it looks really nice good work
Thank you.
I do have two questions regarding the data sources. One, if the numbers line up, would anyone be opposed, to getting all the data from bitcoinity? And not have to deal with the current bitcoincharts.com USD source site, that has invalid credentials that need to be bypassed?
I suppose that if it is easier for you to get it all from one source, then that would be a good idea. I still have not figured out the raw data that is on bitcoinity, and so maybe it would be good to still include a link to the bitcoincharts.com source site so that anyone could look at the raw data there... it is kind of nice to see the raw data... and I personally like to look at it prior to seeing the actual final results for the day (before the day closes).
Two, what should be done if a date entry in the past is missing. Currently there is a JPY day's worth of data missing, far outside the top 100 ratings, from like last year. The script will not run without a number. Should I just repeat the previous day's data, which is what I did this time. Or average the days' data surrounding the missing day? I guess if it's not going to make the top 100 list, and we're not doing any sort of moving average, the 2nd question can be disregarded. But full disclosure is good, so there it is. EDIT: Two days, 2023-08-05, and 2023-08-26. I went back as far as possible over 1200 days, before stopping at the next missing day, 2020-10-12.
You seem to have this part figured out, and if the dates are outside of the top 100, then I cannot imagine how that would end up screwing up anything within the top 100 list..
Thankfully I've also learned to keep backups whenever making a change so rollbacks are possible. It's still a little messed up as we can all see. 