These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE General Discussion

 
  • Topic is locked indefinitely.
 

The Great Ice Mining Interdiction: Not so Great

First post
Author
Previous Topic Next Topic
Victoria Sin
Doomheim
#241 - 2013-08-21 11:49:54 UTC
Given the Calendar API end-points are cached server side, and return loads of duplicates for the same event (yes, it seems broken? https://github.com/ccpgames/esi-issues/issues/216) , can we please have more filtering parameters. I'm open to other ideas, but here is my use-case and problem with ESI currently:

At the moment to get the entries for your corp and alliance for a given time range (e.g. "my corp and alliance event for the last two weeks), one has to:


  1. Call "/characters/{character_id}/calendar/ "
  2. Get the earliest "event_id" from the JSON returned.
  3. Either (walk forward from a known event) or; Walk backwards one request every >5 seconds at the time calling "/characters/{character_id}/calendar/?from_event" for "event_id - (X * )" until you find something at or before your start date.
  4. terate over all responses calling "/characters/{character_id}/calendar/{event_id}/" for each (unique) event_id.
  5. Check the "owner_type" is "corporation" or "alliance"
  6. (Optional) sanity check the owner_id matches the expected corpID / allianceID.
  7. Repeat 4->6 walking forwards on "/characters/{character_id}/calendar/?from_event" for the latest event_id seen until you reach the 'end date' of the range you care about.


Can we please have options to:


  • Suppress duplicate event_id's from "/characters/{character_id}/calendar/"
  • An "EventScope" parameter which allows us to only pull down (Corp | Alliance | Character | All).
  • A "FromDate" (and ideally a "ToDate").
  • A "maxResults" and 'page' parameter allowing a decent number (e.g. 250) events. and iterating over more if the query results in more than maxResults hits.


EXAMPLE:

/v2/characters/{character_id}/calendar/?scopes=Corporation,Alliance&fromDate=2017-02-20&toDate=2017-03-05&maxResults=10&page=2&noDupliates=yes

Would get a only the calendar entries owned by the Corp or Alliance of the character (without duplicate event_ids), between the two dates given, returning entries 11-20 out of however many may have found.
Andski
Science and Trade Institute
Caldari State
#242 - 2013-08-21 11:50:56 UTC
lol, moral high ground in a video game

Twitter: @EVEAndski

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths."    - Abrazzar
Krixtal Icefluxor
INLAND EMPIRE Galactic
#243 - 2013-08-21 11:58:16 UTC
They don't seem to realize that the moral high ground is the world's most expensive real estate.

"He has mounted his hind-legs, and blown crass vapidities through the bowel of his neck."  - Ambrose Bierce on Oscar Wilde's Lecture in San Francisco 1882
S Byerley
The Manhattan Engineer District
#244 - 2013-08-21 12:00:20 UTC
Tippia wrote:
S Byerley wrote:
Still dependent (unless you have some weird definition that's not = 1-P[stolen]); have no problem with P[actually getting the goods] though.
So tell me, how would you write the two in order to control for this dependence on p(stolen)…?


http://www.stat.yale.edu/Courses/1997-98/101/condprob.htm
Tippia
Sunshine and Lollipops
#245 - 2013-08-21 12:10:44 UTC
S Byerley wrote:
http://www.stat.yale.edu/Courses/1997-98/101/condprob.htm

…and you would write the two in order to control for this dependence on p(stolen), how, exactly…?
Malcanis
Vanishing Point.
The Initiative.
#246 - 2013-08-21 12:13:26 UTC
Infinity Ziona wrote:
Tippia wrote:
Infinity Ziona wrote:
You are correct. When suicide ganking the ship is the expendable ammo. A risk is not a risk if the outcome is certain. A risk requires an element of chance.
No. A risk only requires an outcome (usually a cost) and a probability. Just because the probability is 1 doesn't mean the risk goes away.

Yes but lets be realistic here. When the probability falls so low its not a discernible factor there is little to no risk to the person ganking. But if you like than I'll say

"The risk experienced by suicide gankers blowing up miners in barges and so on is so low it could be considered non-existent for anyone not super pedantic".

Does that satisfy you?



it's higher than the risk of taking a peek through a lo-sec gate - which itself seems to to be unacceptably high for the "no risk" brigade.

"Just remember later that I warned against any change to jump ranges or fatigue. You earned whats coming."

Grath Telkin, 11.10.2016
S Byerley
The Manhattan Engineer District
#247 - 2013-08-21 12:15:16 UTC
Tippia wrote:
S Byerley wrote:
http://www.stat.yale.edu/Courses/1997-98/101/condprob.htm

…and you would write the two in order to control for this dependence on p(stolen), how, exactly…?


"If events A and B are not independent, then the probability of the intersection of A and B (the probability that both events occur) is defined by P(A and B) = P(A)P(B|A)."

If you're trying to refer to something else, you'll have to clarify.
Tippia
Sunshine and Lollipops
#248 - 2013-08-21 12:21:37 UTC
S Byerley wrote:
"If events A and B are not independent, then the probability of the intersection of A and B (the probability that both events occur) is defined by P(A and B) = P(A)P(B|A)."
A multiplication, eh? The thing that you said you couldn't do if they were dependent… hmm…

Quote:
If you're trying to refer to something else, you'll have to clarify.
I'm asking you to write the combined probability of the gank succeeding, the item dropping, and the item not getting stolen, and doing it using p[stolen] as your measure of probability.
Frostys Virpio
State War Academy
Caldari State
#249 - 2013-08-21 12:25:41 UTC
Frostys Virpio wrote:
Rhes wrote:
Arduemont wrote:
If a decent group of suicide gankers wants one of your ships dead there is literally nothing you can do about it. It is the only area in Eve where someone gets to PvP risk free.


Except for that whole part where they lose their ships to CONCORD. Roll


If you wish to make people understand that, you are in for a hell of a long time arguing before anyone agree to that.


Called it on page 3. You guys are all wasting your time. I'll see what progress have been made by page 23...
S Byerley
The Manhattan Engineer District
#250 - 2013-08-21 12:41:37 UTC  |  Edited by: S Byerley
Tippia wrote:
S Byerley wrote:
"If events A and B are not independent, then the probability of the intersection of A and B (the probability that both events occur) is defined by P(A and B) = P(A)P(B|A)."
A multiplication, eh? The thing that you said you couldn't do if they were dependent… hmm…


P(B|A) is not dependent on P(A) because it assumes that A happened....

Quote:
Quote:
If you're trying to refer to something else, you'll have to clarify.
I'm asking you to write the combined probability of the gank succeeding, the item dropping, and the item not getting stolen, and doing it using p[stolen] as your measure of probability.


Do it yourself? Maybe you'll learn something.
Logical Chaos
Very Italian People
The Initiative.
#251 - 2013-08-21 12:43:06 UTC
Oh my this thread is so offtopic it is on topic again.
baltec1
Bat Country
Pandemic Horde
#252 - 2013-08-21 13:09:39 UTC
S Byerley wrote:
baltec1 wrote:
So I guess we shouldn't point out the risks of the gank failing or the loot fairy giving the middle finger.


You can if you want; I'd throw it under profit though - deviation maybe under investment (as in, you have to do it for longer to see the same consistency in payout)

How often do your ganks fail?


Often enough for us to scan ships and the space around us to see what defences they have.

Risk is risk, there are no ifs and buts about this, a gank can and will fail for any number of reasons.
baltec1
Bat Country
Pandemic Horde
#253 - 2013-08-21 13:10:58 UTC
Frostys Virpio wrote:


Called it on page 3. You guys are all wasting your time. I'll see what progress have been made by page 23...


Would you like another dead orca to feast upon while you wait?
Tippia
Sunshine and Lollipops
#254 - 2013-08-21 13:13:07 UTC
S Byerley wrote:
P(B|A) is not dependent on P(A) because it assumes that A happened.
…so much like it being stolen (or not), assuming that it has dropped, you mean?

Quote:
Do it yourself?
No. How about instead you prove that you actually know what you're talking about rather than presenting links that only show that you are aware of google.

So, how would you write the two in order to control for the dependence of p(stolen)…?
Prince Kobol
#255 - 2013-08-21 13:13:49 UTC
If you insist on fitting for yield then going afk then you deserved to be ganked.

You want to avoid being a target, fit a decent tank, don't go afk and mine next to the idiot who is fitted for yield and is afk, then laugh as he is blown to tiny little pieces.

Its really is that simply
Ozmodan
Pator Tech School
Minmatar Republic
#256 - 2013-08-21 13:25:27 UTC
If a Mac lets 4 catalysts pop him he was asleep at the wheel!
Malcanis
Vanishing Point.
The Initiative.
#257 - 2013-08-21 14:06:19 UTC
Since we're agreed that the loss of the gankers ship to CONCORD is not a risk and serves no purpose in deterring ganking, shall I go ahead and ask CCP to remove this useless mechanic? It's rather demanding of system resources, and takes up development and code maintenance resources that could be applied on more useful mechanics

"Just remember later that I warned against any change to jump ranges or fatigue. You earned whats coming."

Grath Telkin, 11.10.2016
PhatController
Deep Core Mining Inc.
Caldari State
#258 - 2013-08-21 14:22:06 UTC
For the average ganker, yes the changes should make it harder. But someone on goons scale it shouldn't be hard. Alt in each ice system checks for belts. When one is up roaming gang of gankers flys to system and causes havoc. To be fair, if they are doing there job correctly, the belts will be up 24/7 :P

It's early days yet, it could be awhile before we see how this progresses.
Frostys Virpio
State War Academy
Caldari State
#259 - 2013-08-21 14:34:31 UTC
baltec1 wrote:
Frostys Virpio wrote:


Called it on page 3. You guys are all wasting your time. I'll see what progress have been made by page 23...


Would you like another dead orca to feast upon while you wait?


I might go work on getting my feast tonight instead of relying on other to provide it. If I get time for EVE that is...
Frostys Virpio
State War Academy
Caldari State
#260 - 2013-08-21 14:38:59 UTC
PhatController wrote:
For the average ganker, yes the changes should make it harder. But someone on goons scale it shouldn't be hard. Alt in each ice system checks for belts. When one is up roaming gang of gankers flys to system and causes havoc. To be fair, if they are doing there job correctly, the belts will be up 24/7 :P

It's early days yet, it could be awhile before we see how this progresses.


They don't even need to do that. They can let the miners mine all of it then gank the haulers/freighters and it would choke the supply chain just as good. Mined or not, the ice is useless in osmon for example.