[BPIP] Bitcointalk Public Information Project [Back in Action]

[jamyr]

A few steps away from tangible at the moment, but I do intend to see these printed at some point. Smiley

Like a real life bitcointalk badge or identification card, woohoo!!

If its an ID it should have a :

In case of bitcointalk emergency contact : Satoshi
hahahaha

You could see some samples here (https://bitcointalk.org/index.php?topic=5300899.msg55851705#msg55851705).
Although, most recent sample would be here (https://bitcointalk.org/index.php?topic=5319417.msg56481846#msg56481846).

If I am not mistaken I saw this before but did not really read it thoroughly, I just remember seeing cards with bitcointalk "celebrities".
(if I am mistaken then it must be a card that looked like pokemon cards or Yu-gi-oh inspired cards)

Thanks for the suggestion, it should be possible to capture the date when the parsers have detected a signature changing. But since there is no official log displaying temporary bans, I couldn't guarantee it to be the exact time, but it should be close enough to be able to determine how many days of a ban it might be. I can add this to the work list and we'll take a look.

I didn't find anything up my sleeves to help, (but I found something hairy in my underarms. lol),

". .I couldn't guarantee it to be the exact time"

Even scientists can only give approximation to when humans started to inhibit this planet, they say "approximately . . yada yada million years. . . ago", so close enough is <3 ,


edit

I just realized this information of when a signature is updated can be beneficial for the services section. For example, you applied for a signature campaign but doesn't get a confirmation if you are accepted or not.
So managers would have an idea when you start wearing the signature and can easily compare it to your latest post.

[1714119639]

1714119639

[LoyceV]

Bug?

No. CanaryInTheMine just dropped out of DT1, see DT Change Log:

icopress DT1 (-1) distrusts CanaryInTheMine DT1 (-1)

And with that, his Trust score dropped from +202 / =0 / -0 to +21 / =0 / -0. If you click "REFRESH PROFILE" in BPIP it should normalize.
That's probably related to my post here.

[condoras]

I also take a look today at my profile and I saw a difference in my Trust ratings as well.
I can't explain it but I thought to report it here so you guys look at it. Btw, I refresh the page, and still the same...

[Timelord2067]

I also take a look today at my profile and I saw a difference in my Trust ratings as well.
I can't explain it but I thought to report it here so you guys look at it. Btw, I refresh the page, and still the same...

This is how I see your profile (with the ;dt toggle on) and BPIP scores at the moment:





A couple of people have been ... shall we say tweeking their default trust lists which has had an impact on quite a few other users' scores.

[condoras]

A couple of people have been ... shall we say tweeking their default trust lists which has had an impact on quite a few other users' scores.

I read the post of LoyceV about CITM but I didn't think that it will have such an impact. It seems that I thought wrong.
Can you tell me what these marks/ signs are and what they mean?

[LoyceV]

Can you tell me what these marks/ signs are and what they mean?

See [β] BPIP Extension: user info & extra features add-on/extension, Firefox/Chrome.

[Bitcoin_Arena]

Does BPIP track unbanned members or this user was never actually got unbanned?

Over, 2 years ago Cellard was banned but he made a ban appeal later.

According to this statement, hilariousandco reduced the ban to just 60 days but BPIP shows as though the user is still banned as of today, 2 years later.

I changed the ban to 60 days + 1 year sig ban seeing as it's a post from years ago and you seem to have contributed decently since.

Unfortunately, Cellard never posted again so it hard to know if he was actually unbanned (BPIP not updating unban info) or still is banned.

[suchmoon]

Unfortunately, Cellard never posted again so it hard to know if he was actually unbanned (BPIP not updating unban info) or still is banned.

Cellard is no longer banned but unfortunately unbans are not shown in modlog or anywhere else where we can automatically scrape it. We do remove bans when banned users start posting again but as you noticed Cellard hasn't posted since the ban.

[Timelord2067]

There's something strange going on with the Trust Log page:

Here are two screen shots I've just taken the second is after I refreshed the page, notice the time index of the first two entries.  Each time the page is refreshed, either theymos or OgNasty's entries are displayed immediately after the first "new" entry.

This has been occurring since the most recent change in the DT1 appointments.

[LoyceV]

Each time the page is refreshed, either theymos or OgNasty's entries are displayed immediately after the first "new" entry.

I get JollyGood when I refreshed, but that happened only once.

Is there a reason the Search box disappeared from that page?

[suchmoon]

Each time the page is refreshed, either theymos or OgNasty's entries are displayed immediately after the first "new" entry.

I get JollyGood when I refreshed, but that happened only once.

It's sorted by time, but when multiple records have the same timestamp the sort order within that group is undefined so it can be kinda random. I'd like to say I'll fix it but this might take a while

Is there a reason the Search box disappeared from that page?

It's been replaced by these filter options and now it works across the whole history, not just the last 1000 records or whatever the limit used to be:

[LoyceV]

Thanks, I would have never guessed to click there.

[ibminer]

Thanks, I would have never guessed to click there.

There is also a 'report options' button in the middle of the screen which pulls out that side menu as well.. I was hopeful putting that button where the filter box used to be would draw attention to it, but I probably should have moved it under the page(s) display.  

It's sorted by time, but when multiple records have the same timestamp the sort order within that group is undefined so it can be kinda random. I'd like to say I'll fix it but this might take a while

I can try to work this into the development site over the next few weeks, should just be a secondary sort that needs to be added into the query.

I assume it makes sense to use the username, alphabetical, as the secondary sort.

[Timelord2067]

It's sorted by time, but when multiple records have the same timestamp the sort order within that group is undefined so it can be kinda random. I'd like to say I'll fix it but this might take a while

I can try to work this into the development site over the next few weeks, should just be a secondary sort that needs to be added into the query.

I assume it makes sense to use the username, alphabetical, as the secondary sort.

Is it really necessary to restate information that could be months, if not years old?  By restating that a user trusts/distrusts another user all you are doing is filling your own logfiles with endless pages of repetitive information.

Surely stating a person has been made, or removed from, DT1 should be enough information, otherwise it's going to appear as though DT1 users can't make up their minds repeatedly trusting/distrusting when in fact it didn't happen.

[ibminer]

Is it really necessary to restate information that could be months, if not years old?  By restating that a user trusts/distrusts another user all you are doing is filling your own logfiles with endless pages of repetitive information.

Surely stating a person has been made, or removed from, DT1 should be enough information, otherwise it's going to appear as though DT1 users can't make up their minds repeatedly trusting/distrusting when in fact it didn't happen.

Assuming you are referring to the DT1 lottery/shuffle each month and the repetitive entries that occur because of it, those repetitive entries are not intentional, they are more of a side effect on how the parser is currently reading changes to DT1 inclusion/exclusion lists. I'm hopeful this will get cleaned up at some point, assuming we find a good way to isolate those monthly changes to either identify & condense them.. or maybe hide them?.. but no, those particular repetitive entries each month are not necessary.

I do think the historical reference of DT1 trust changes can be relevant, so I think it still makes sense to save the history, in general.

[Timelord2067]

Hello again,

Just a quick question if I may:  What is the highest number that BPIP scores go up to?  I was always under the impression it was just one thousand, however, I've noticed recently UID's with much higher scores.

E.G.

I also figured out the ranking of users with the same score is based on their registration date instead of giving each user the same score/rank giving earlier registered users a nudge up the totem pole...

[ibminer]

It's sorted by time, but when multiple records have the same timestamp the sort order within that group is undefined so it can be kinda random. I'd like to say I'll fix it but this might take a while

I can try to work this into the development site over the next few weeks, should just be a secondary sort that needs to be added into the query.

I assume it makes sense to use the username, alphabetical, as the secondary sort.

Instead of adding anything, switching the sorting to the 'id' itself seemed to be the best solution. Assuming everything is as it should be, I believe this should always still coincide with the timestamp, but it keeps the /TrustLog page less jumpy. It was a simple switch, so I've made this modification live as well. Might have to switch it back if I did something stupid.. but it seems more stable now. 

Hello again,

Just a quick question if I may:  What is the highest number that BPIP scores go up to?  I was always under the impression it was just one thousand, however, I've noticed recently UID's with much higher scores.

Regardless of what is displayed on the site, because ranks are based on calculations, these calculations could be applied to any user.. but it does require more resources, and isn't really necessary to run the calculations on everyone for the website. Only the top 1,000 are currently displayed/calculated on the website, but there are intentions on increasing this to maybe 10,000 once we can do it across the board. The extension specifically calculates the recognition values/ranks past 1,000 and lists that as the BPIP rank.

I also figured out the ranking of users with the same score is based on their registration date instead of giving each user the same score/rank giving earlier registered users a nudge up the totem pole...

The "Most Recognized" rank in your example does not intentionally do any sorting based on the registration date. The only calculation done is listed at the top of the most recognized page.

However, I'm guessing because SQL processes the tables with profile ID's in order, the lower/older userIDs will probably get processed first, which would get them ranked first.

It's not intentional, but as it stands now, I don't see a huge issue with this personally. It could be something adjustable in the future but I'm not sure it is a huge issue at this point.

[Timelord2067]

...

Your answer has confused my two separate questions; treating them as one and the same. 

But, thank-you for at least in part attempting to answer the second question as it in part answers questions put to you more than twelve months ago by others, not just myself.

[LoyceV]

At the risk of being slightly off-topic: Does BPIP also get visited by agressive scraping bots? Since yesterday, I've banned about 15,000 IP addresses from accessing loyce.club because they were scraping the same data directory with many downloads per second. The fact that they use so many IP addresses makes it annoying to deal with and a very agressive way to collect data.
Open IP tracker shows the IPs come from all over the world and are owned by different organisations.

[ibminer]

At the risk of being slightly off-topic: Does BPIP also get visited by agressive scraping bots?

Not sure.. I'm sure we have bots scraping pages, but I'm not sure how many or how aggressive they may be. I'm guessing our hosting provider is doing some level of filtering on their end, but again I'm not sure how involved it is or whether suchmoon does any monitoring of these types of things on his end, GUI access may have a few more options to monitor this type of thing more effectively. I could probably try digging through long text-based log files and trying to analyze them, but who has time for that??