GLBSE 2.0 open for testing

[Nefario]

Nefario,

There appears to be a bug in the way your script decides what color to make a stock in the assets list.  MergedMining shares are currently showing up as red, which seems to indicate they have dropped in value.  However, looking at the list of share sales, you can see that shares are currently at a monthly high.  So it should be green, right?

http://twitter.com/#!/search/realtime/mergedmining

I'll investigate this, the way it currently works is the color depends on the price of the last trade compared to the one before it.

So for example, if the latest trade was for more than the one before it then it's green.

[OgNasty]

Nefario,

There appears to be a bug in the way your script decides what color to make a stock in the assets list.  MergedMining shares are currently showing up as red, which seems to indicate they have dropped in value.  However, looking at the list of share sales, you can see that shares are currently at a monthly high.  So it should be green, right?

http://twitter.com/#!/search/realtime/mergedmining

I'll investigate this, the way it currently works is the color depends on the price of the last trade compared to the one before it.

So for example, if the latest trade was for more than the one before it then it's green.

That doesn't appear to be working correctly. Not a big deal, just reporting a bug. Also, a more traditional approach would be using the prior day's closing price instead of the last trade. Great work on 2.0 so far.

[BinaryMage]

It may be indeed useless, or it may be not. Nice to hear your opinion on the matter. Mine is not so certain.

You don't need to thank me for my opinion, I'll give it whether you like it or not.
I'm not saying that I think Litecoin is a bad investment; I'm saying that from a purely technical standpoint it has nothing to differentiate itself from Bitcoin except the hash algorithm difference, which lends itself to weakness. (An algorithm inefficient on standard hardware opens up the currency to an easy 51% attack by custom hardware designed to use that particular algorithm) If you have evidence against that, please say so.

Lots of people mines with CPUs. You may not consider this "serious" mining, but it happens, and at least for those people, there should be some value in doing it.

"No one" was hyperbole, I understand that people do actually mine with CPUs.
"There should be some value in doing it"? Value is not brought in by miners; miners are brought in by value. The existence of CPU miners doesn't necessarily mean it's a good idea to create a currency that allows them to feasibly profit from CPU mining.

More miners means more people, more human resources if you prefer, doing something they found valuable. That's the way anything becomes valuable.

It can, I agree, but not in every case. I mined Litecoins for awhile because it made financial sense. I thought the concept was useless and I sold every Litecoin I mined immediately. If you think that adds value to the currency, please explain.

In fewer words, it's the GLBSE, and there's a darn good reason for that.

So, just because bitcoin is hardcoded on the name, we should make separate projects for any other currency? Ok, fair with me, but I think it would be almost trivial to integrate other cryptocurrencies on the existing infrastructure, and it certainly would aggregate value to it.

Don't over-analyze that statement, it was just a tl;dr summary. Could other currencies be integrated? Yes. Would it aggregate value? I'm not so sure.

I know that all of us who are longing bitcoin would like that someday all the value in the universe would be measured in bitcoin, but sadly this isn't going to happen. There will always be lots of units of measure and people will be choosing arbitrarily between them.

And I think that a more complex cryptocurrency ecosystem would be beneficial for bitcoin as a whole. I tend to see it as in the free software ecosystem. Just because linux exists and is the kernel chosen by the big majority of people, that doen't meant that we should abandon projects like gnu/hurd, *bsd and every other competing kernel. On the contrary, the existence of alternatives and competition makes the big projects even more valuable.

Oh, I certainly hope Bitcoin doesn't get to that point, at least in it's current state - it's far too vulnerable to attack. I agree entirely with the idea of competition, but your analogy is misfitting. *BSD and gnu/hurd provide advantages that Linux does not. (Trust me, I use the former. FreeBSD is more secure, often faster, and significantly more stable than Linux in my humble opinion.) However, I fail to see (and if there are any please enlighten me) any advantages Litecoin provides relative to Bitcoin. Competition is incredibly valuable - spin-off schemes with no advantages are not.

[Nefario]

That doesn't appear to be working correctly. Not a big deal, just reporting a bug. Also, a more traditional approach would be using the prior day's closing price instead of the last trade. Great work on 2.0 so far.

Define closing price when the exchange never closes, would it be 00:00 GMT or something along those lines?

[OgNasty]

That doesn't appear to be working correctly. Not a big deal, just reporting a bug. Also, a more traditional approach would be using the prior day's closing price instead of the last trade. Great work on 2.0 so far.

Define closing price when the exchange never closes, would it be 00:00 GMT or something along those lines?

Good point.  This leaves room for some innovation.    Perhaps it could be based off of a running 48 hour average?

[Sukrim]

I'd go with GMT/UTC 00:00 as "new day" personally... any other time is as arbitrary anyways but this allows for some "chunking".

On a different note: I went through the hassle of claiming one of my 2 GLBSE 1.0 accounts, but getting an error that "Claim code: xxxXXXxxxXXXxxx has not been found, please check for typo's[sic!]" even though it's copy-pasted from the 1.0 account popup (32 characters strong). Is there some wait time until the systems are synced or is this a bug?

Also I plan to claim my other account too and transfer it's assets over to the first one (likely via a throwaway account that claims account #2 and just transfers assets) - but it would be nice to have the history too, so if it is at all possible to claim 2 old accounts into 1 new account, please tell me.

[mila]

both is used, different sites, different implementations
gox uses 24h avg
the bitcoincharts reset at midnight

[Nefario]

I'd go with GMT/UTC 00:00 as "new day" personally... any other time is as arbitrary anyways but this allows for some "chunking".

On a different note: I went through the hassle of claiming one of my 2 GLBSE 1.0 accounts, but getting an error that "Claim code: xxxXXXxxxXXXxxx has not been found, please check for typo's[sic!]" even though it's copy-pasted from the 1.0 account popup (32 characters strong). Is there some wait time until the systems are synced or is this a bug?

Also I plan to claim my other account too and transfer it's assets over to the first one (likely via a throwaway account that claims account #2 and just transfers assets) - but it would be nice to have the history too, so if it is at all possible to claim 2 old accounts into 1 new account, please tell me.

This is a bug, one that I'm working on, a couple of people have had this problem.

Yes it's possible to claim more than one GLBSE1.0 account into a single GLBSE2.0 account.

[Sukrim]

Nice, so I'll just note down the code and try again in a couple of days...

Happy easter by the way!

[Brian DeLoach]

I understand about Yubikey - but it's cost free to implement would just be an added feature for GLBSE.  I can show you how to implement it without buying a Yubikey if you want, it's actually pretty straight forward.  There's no real drawback to offering it, once you get it setup, it pretty much is hands off at that point.

Any final decision regarding yubikeys? It would be a nice feature.

[OgNasty]

I'd go with GMT/UTC 00:00 as "new day" personally... any other time is as arbitrary anyways but this allows for some "chunking".

The "new day" works great on exchanges with high volume and a typical news cycle.  After giving this some thought, I don't think the "new day" approach would be typically beneficial for the GLBSE.  I don't even think my "48-hour average" approach would really capture enough data to show whether or not a stock is increasing or decreasing in value.  If it were my decision, I think I would use a 7-day running average as a base to judge whether or not a stock is increasing or decreasing.  That should capture enough trades to actually display a useful metric (green/yellow/red).  As the exchange gets more popular and volume increases, this time could be reduced.

[teek]

I'd go with GMT/UTC 00:00 as "new day" personally... any other time is as arbitrary anyways but this allows for some "chunking".

The "new day" works great on exchanges with high volume and a typical news cycle.  After giving this some thought, I don't think the "new day" approach would be typically beneficial for the GLBSE.  I don't even think my "48-hour average" approach would really capture enough data to show whether or not a stock is increasing or decreasing in value.  If it were my decision, I think I would use a 7-day running average as a base to judge whether or not a stock is increasing or decreasing.  That should capture enough trades to actually display a useful metric (green/yellow/red).  As the exchange gets more popular and volume increases, this time could be reduced.

+1  7 day is what it needs for now

[mila]

is there an exploitable bug the forum.glbse ?
some dude just posted a bazilion spam posts that should not be possible
(it if the 2min limit would be still applied)

[Nefario]

is there an exploitable bug the forum.glbse ?
some dude just posted a bazilion spam posts that should not be possible
(it if the 2min limit would be still applied)

Nope, just someone who patiently took the time to make all those posts which are now deleted.

[mila]

is there an exploitable bug the forum.glbse ?
some dude just posted a bazilion spam posts that should not be possible
(it if the 2min limit would be still applied)

Nope, just someone who patiently took the time to make all those posts which are now deleted.

thanks. that was lightning fast!
// it was probably scripted (or waste of time to post manually)

edit: btw any news on trade history (list trades per asset)?

[Sukrim]

I'd go with GMT/UTC 00:00 as "new day" personally... any other time is as arbitrary anyways but this allows for some "chunking".

The "new day" works great on exchanges with high volume and a typical news cycle.  After giving this some thought, I don't think the "new day" approach would be typically beneficial for the GLBSE.  I don't even think my "48-hour average" approach would really capture enough data to show whether or not a stock is increasing or decreasing in value.  If it were my decision, I think I would use a 7-day running average as a base to judge whether or not a stock is increasing or decreasing.  That should capture enough trades to actually display a useful metric (green/yellow/red).  As the exchange gets more popular and volume increases, this time could be reduced.

+1  7 day is what it needs for now

something like this?

Code:

| Asset code | Last trade | 7 day average | 7 day trend line | 
--------------------------------------------------------------
| ABC123     |   13@3.37  |          3.03 | ____________/\_/ | <-- green background, because 3.37 > 3.03

Maybe there could be even something like "average of the last 50 trades" or "average of the last x trades using 10% of shares emmitted"?
The last metric would mean that this is the average price per share of the latest 20 shares traded, if your IPO was 200 share. This would eliminate the time window completely but could be too hectic for high volume trades maybe...

[mila]

the graph in asset view is rounding the trade price to 3 decimal places
bug or feature?

and could be volume added to the graph as a second indicator in the picture, please?

[Nefario]

the graph in asset view is rounding the trade price to 3 decimal places
bug or feature?

and could be volume added to the graph as a second indicator in the picture, please?

The rounding is a problem using Google graphs, the data being sent is not rounded, but Google charts seems to round them out.

We'll be moving to much better graphs, that also include trade volume as well.

[rdponticelli]

Don't you thing it could be nice to keep the testing site working? It could be useful to those planing to launch an asset, to become more acquainted with the functionality, before doing it for real. Maybe it would be too much overhead to you?

[Brian DeLoach]

I understand about Yubikey - but it's cost free to implement would just be an added feature for GLBSE.  I can show you how to implement it without buying a Yubikey if you want, it's actually pretty straight forward.  There's no real drawback to offering it, once you get it setup, it pretty much is hands off at that point.

Any final decision regarding yubikeys? It would be a nice feature.

Bump.